1VATI
WATER POLLUTION CONTROL RESEARCH SERIES • 16100 EXH 11/71
INTERNATIONAL SYMPOSIUM
ON
WATER POLLUTION CONTROL
IN
COLD CLIMATES
ENVIRONMENTAL PROTECTION AGENCY0RESEARCH AND MONITORING
-------
International Symposium on
WATER POLLUTION CONTROL
IN
COLD CLIMATES
held at the University of Alaska
July 22-24, 1970
Sponsored by
INSTITUTE OF WATER RESOURCES
UNIVERSITY OF ALASKA
and
FEDERAL WATER QUALITY ADMINISTRATION
EDITORS
R. Sage Murphy David Nyquist
Director, Institute of Assistant Professor of
Water Resources Water Resources
TECHNICAL EDITOR
Paul W. Neil
For sale by the Superintendent of Documents, U.S. Government Printing Office
Washington, D.C. 20402 - Price $2.60 (paper cover)
Stock Number £501-0208
-------
INTRODUCTION
The first plans for this symposium were begun by the senior editor during the winter of 1967.
Alaska was about to become the focal point for petroleum interests as a consequence of the
enormous oil discoveries made by the Atlantic-Richfieid and Humble Oil joint venture at Prudhoe
Bay. Coincident with an economic boom in this sparsely-settled region, federal, state, and private
organizations instituted more stringent water pollution control criteria. The combined events
created an apparent technology gap relative to water pollution control in cold climates. The
engineering and scientific bases for waste treatment and receiving stream criteria were almost
exclusively dependent upon temperate experiences. Scattered journal articles about the North
were in evidence, but the majority seemed repetitious and contributed little toward a true
understanding of the problems involved.
It became obvious that more experience in northern applications was available in the circumpolar
countries than in the temperate areas of the United States. A symposium of international scope
was therefore envisioned, with representatives of all the circumpolar nations presenting papers.
With the exception of the Soviet Union and Iceland, all countries participated. The eighteen papers
and two reviews in this volume were selected from approximately sixty invited abstracts.
Persons living or interested in the Far North need no statistics to convince them of the area's
potential. It is not the purpose of this introduction to expound on the virtues of the North;
however, it is our firm belief that the Arctic and sub-Arctic will be focal points for economic and
resource development, and for available open space for the rest of the world in the imminent
future. Some implications of this have already become manifest.
Should such development indeed occur, it would certainly be imperative that workers in the
circumpolar countries become entirely familiar with the work done in each.
Therefore, besides descriptions of engineered works and scientific studies, the meeting permitted
many the opportunity to personally meet their international counterparts. Since such contact is
often more important in the long run than the formal papers, hopefully a continuing exchange of
information between the various groups concerned with water pollution control will evolve.
The first meeting of such a group is usually difficult. A wide range of disciplines and organizations
were involved. The most advanced waste treatment technology is of no use unless the limits of
materials able to be discharged into the receiving waters are understood; consequently, invited
papers were evenly split between "waste treatment technology" and "effects of wastes upon Far
Northern receiving waters." Thus, an attempt was made to create an atmosphere where engineers
and biologists could communicate. In this endeavor partial success was achieved. If subsequent
meetings are held in the future, it is anticipated that they will be devoted to single, more specific
topics than was our general convocation. Indications are that such groups will be meeting in
Alaska, Canada, and Scandanavia in the very near future.
The success of any meeting depends upon a large number of people. We thank the authors for their
thoughtful manuscript preparation. The secretarial staff and graduate research assistants of the
University of Alaska's Institute of Water Resources unselfishly devoted many hours handling the
thankless details both in preparation for, and during, the symposium. The banquet talk, which is
not included in these papers, was given by Mr. Geoffrey Larminie, area manager for British
i ii
-------
Petroleum (BP Alaska, Inc.). His talk about the Prudhoe Bay oil fields, how they will be and are
being managed, and the implications concerning water pollution control will be remembered by all
in attendance.
No meeting is successful without financial backing. The majority of funds for the symposium were
provided by the Federal Water Quality Administration, U. S. Department of the Interior (now the
Environmental Protection Agency), through grant number 16100 E X H. We are particularly
indebted to Mr. Richard Latimer, Director of FWQA's Alaska Water Laboratory, and Mr. William
Caw ley of FWQA's Washington, D. C. staff, for their efforts in making the meeting a success. The
Office of Water Resources Research, U. S. Department of the Interior, and the State of Alaska,
through their funding of our Institute, contributed significantly, though indirectly, to this
meeting. Without their support the symposium could not have been attempted.
The editors sincerely hope this volume will help to bring together some of the existing practices
and theories concerned with water pollution control in cold climates.
R. Sage Murphy
David Nyquist
IV
-------
ACKNOWLEDGMENT
In appreciation of financial contributions without which the Symposium could not have been
convened.
Atlantic Richfield Company
BP Alaska, Incorporated
Humble Oil & Refining Company
Marathon Oil Company
Texaco. Incorporated
-------
TABLE OF CONTENTS
RECEIVING WATERS
SYNOPTIC STUDY OF ACCELERATED EUTROPHICATION IN LAKE
TAHOE - AN ALPINE LAKE 1
Charles R. Goldman, Professor of Zoology, Institute of Ecology. University
of California, Davis, California, U.S.A.
Gerald Moshiri, Institute of Ecology, University of California, Oavis,
California, U.S.A.
Evelyne de Amezaga, Institute of Ecology, University of California, Davis,
California, U.S.A.
THE SOUTH BASIN OF LAKE WINNIPEG • AN ASSESSMENT OF POLLUTION 22
Jo-Anne M. E. Crowe, Pollution Biologist, Fisheries Branch, Department of
Mines and Natural Resources, Winnipeg, Manitoba, Canada
EUTROPHICATION IN SOME LAKES AND COASTAL AREAS IN FINLAND,
WITH SPECIAL REFERENCE TO POLYHUMIC LAKES „ 48
Pasi O. Lehmusluoto, Assistant in the Department of Limnology,
University of Helsinki, Helsinki, Finland
THE RECOVERY PROCESS OF A LAKE WHICH RECEIVED WASTEWATER
FROM AN ORE DRESSING PLANT 61
Bengt Ahling, Research Engineer, Swedish Water and Air Pollution
Research Laboratory. Stockholm, Sweden
DEPLETION OF OXYGEN BY MICROORGANISMS IN ALASKAN RIVERS
AT LOW TEMPERATURES 71
Ronald C. Gordon, Research Microbiologist, Alaska Water Laboratory,
Federal Water Pollution Control Administration, College, Alaska, U.S.A.
PREDICTION OF DISSOLVED OXYGEN LEVELS IN THE SOUTH
SASKATCHEWAN RTVER IN WINTER 96
Robert C. Landine, Assistant Professor, Department of Civil Engineering,
University of New Brunswick, Fredericton, New Brunswick, Canada
VI I
-------
POLLUTION - A BIOLOGICAL STUDY OF SOME RECEIVING WATERS IN HOKKAIDO 113
Matsunae Tsuda, Professor of Zoology, Zoological Institute, Nara Women's
University, Nara, Japan
Toshiharu Watanabe, Zoological Institute. Nara Women's University, Nara,
Japan
Kozo Tani, Zoological Institute, Nara Women's University, Nara, Japan
CHEMICAL EFFECTS OF SALMON DECOMPOSITION ON AQUATIC ECOSYSTEMS 125
David C. Bricked, Graduate Research Assistant, Institute of Marine
Sciences, University of Alaska, College, Alaska, U.S.A.
John J. Goering, Professor of Marine Sciences, Institute of Marine Sciences,
University of Alaska, College, Alaska, U.S.A.
PHOSPHORUS BINDING MECHANISMS DURING SELF-PURIFICATION OF
POLLUTED LAKES 139
Jan Werner, Research Chemist, Swedish Water and Air Pollution Research
Laboratories, Stockholm, Sweden
CRITICAL REVIEW OF PAPERS ON RECEIVING WATERS 153
Peter A. Krenkel, Chairman and Professor, Department of Environmental
and Water Resources Engineering, Vanderbilt University, Nashville,
Tennessee, U.S.A.
TREATMENT PROCESSES
THE INFLUENCE OF TEMPERATURE ON THE REACTIONS OF THE
ACTIVATED SLUDGE PROCESS 164
Pal Benedek, Head of Water Quality and Technology Department, Research
Institute for Water Resources Development, Budapest, Hungary
Peter Farkas, Senior Research Biologist, Department of Water Quality and
Technology, Research Institute for Water Resources Development,
Budapest, Hungary
VI I I
-------
TEMPERATURE EFFECTS ON BIOLOGICAL WASTE TREATMENT PROCESSES 180
W. Wesley Eckenfelder, Jr., Professor of Environmental and Water
Resources Engineering, Vanderbilt University, Nashville, Tennessee, U.S.A.
Andrew J. Englande, Jr., Research Assistant, Department of Environmental
and Water Resources Engineering, Vanderbilt University, Nashville,
Tennessee, U.S.A.
EVALUATION OF AERATED LAGOONS AS A SEWAGE TREATMENT
FACILITY IN THE CANADIAN PRAIRIE PROVINCES 191
Archie R. Pick, Research Engineer, Water Works and Waste Disposal
Division, The Metropolitan Corporation of Greater Winnipeg, Winnipeg,
Manitoba, Canada
George E. Burns, Engineer of Design, Water Works and Waste Disposal
Division, The Metropolitan Corporation of Greater Winnipeg, Winnipeg,
Manitoba, Canada
Dick W. Van Es, Engineer of Sewage Disposal, Water Works and Waste
Disposal Division, The Metropolitan Corporation of Greater Winnipeg,
Winnipeg, Manitoba, Canada
Richard M. Girling, Assistant Engineer of Design, Water Works and Waste
Disposal Division, The Metropolitan Corporation of Greater Winnipeg,
Winnipeg, Manitoba, Canada
DESIGN CONSIDERATIONS FOR EXTENDED AERATION IN ALASKA 213
Sidney E. Clark, Acting Chief, Cold Climate Research Program, Alaska
Water Laboratory, Federal Water Quality Administration, College, Alaska,
U.S.A.
Harold J. Coutts, Research Engineer, Cold Climate Research Program,
Alaska Water Laboratory, Federal Water Quality Administration, College,
Alaska. U.S.A.
Conrad Christiansen, Research Engineer, Cold Climate Research Program,
Alaska Water Laboratory, Federal Water Quality Administration, College,
Alaska, U.S.A.
CHEMICAL TREATMENT OF MECHANICALLY AND BIOLOGICALLY
TREATED WASTEWATER 237
Arne Rosendahl, Sanitary Engineer, Norwegian Institute for Water
Research, Oslo, Norway
IX
-------
BIOLOGICAL AND CHEMICAL WASTE TREATMENT EXPERIMENTS IN
FAR NORTHERN SWEDEN 252
Peter Balmer, Consulting Engineer, Allmanna Ingenjorsbyran AB,
Stockholm, Sweden
BIOLOGICAL SEWAGE TREATMENT IN A COLD CLIMATE AREA 263
Keichi Koyama, Associate Professor, Department of Sanitary Engineering,
Hokkaido University. Sapporo, Japan
Shigeo Terashima, Professor, Department of Sanitary Engineering,
Hokkaido University, Sapporo, Japan
Yasumoto Magara, Instructor, Department of Sanitary Engineering,
Hokkaido University, Sapporo, Japan
MICROBIOLOGIC INDICATORS OF THE EFFICIENCY OF AN AERATED,
CONTINUOUS-DISCHARGE, SEWAGE LAGOON IN NORTHERN CLIMATES 286
John W. Vennes. Professor of Microbiology, School of Medicine. University
of North Dakota, Grand Forks, North Dakota, U.S.A.
Otmar 0. Olson, School of Medicine, University of North Dakota, Grand
Forks. North Dakota, U.S.A.
DISINFECTION AND TEMPERATURE INFLUENCES 312
Cecil Chambers, Research Microbiologist, Biological Research Program,
Advanced Waste Treatment Research, Federal Water Pollution Control
Administration, Cincinnati, Ohio, U.S.A.
Gerald Berg. Chief, Virology Section, Waste Identification and Analyses
Activities, Advanced Waste Treatment Research, Federal Water Pollution
Control Administration, Cincinnati. Ohio, U.S.A.
CRITICAL REVIEW OF PAPERS ON TREATMENT PROCESSES 329
Karl Wuhrmann, Professor, Swiss Federal Institute of Technology, Zurich,
Switzerland
-------
SYNOPTIC STUDY OF ACCELERATED EUTROPHICATION
IN LAKE TAHOE-AN ALPINE LAKE
Charles R. Goldman, Gerald Moshiri
and Evelyne de Amezaga
INTRODUCTION
Lake Tahoe, located at an elevation of 6,229 feet near the crest of the Sierra-Nevada in a graben
basin, is among the clearest alpine lakes in the world. Secchi measurements over 35 meters have
been recorded and the lake still has an extinction coefficient between 0.047 and 0.061 rn"1. Its
watershed of 800 Km2 is small in comparison to its surface of 499 Km2. The average depth of
Tahoe is 313 meters and it contains 156 Km3 of water with a retention time of about 700 years
(Fig. 1).
With the exception of the early observations of Le Conte (1883a, 1883b, 1884) and the
limnological reconnaissance of Juday (1907), Kemmerer et al. (1923), and Hutchinson (1937), the
lake has received little limnological study until the last ten years. Geological studies of the lake
have included fathometry of the basin as well as studies of sediment distribution (Court, Goldman,
and Hyne, in press). The general trophic state of Tahoe may be classified as very oligotrophic as
compared with a variety of other lake types (Goldman, 1967, 1968). Attached aquatic plants are
found to a depth of 100 meters (Frantz and Cordone, 1967) and the crayfish Pacifasticus
leniusculus is extremely abundant in the littoral zone (Abrahamsson and Goldman, 1970).
.av« Rock
Uppw Truckw River
FIGURE 1 Lake Tahoe, California-Nevada
1
-------
Culture experiments show that the lake is deficient in available iron, that the interaction between
nitrogen and phosphorus in Tahoe is very complicated, and that the various tributaries of the lake
differ greatly in their stimulation of algal growth (Goldman, 1964; Goldman and Armstrong, 1969;
Goldman, Tunzi, and Armstrong, 1969).
The variation in productivity in the lake was first documented during two synoptic studies in 1962
(Goldman and Carter, 1965). A second series of three synoptics was run in 1967, and a third series
in 1968 which are reported here.
During the last decade the Tahoe basin, an important recreational area of the Sierra-Nevada, has
undergone a very rapid increase in both resident and visitor population, and despite its great
volume of low nutrient concentration, the measured increase in primary productivity is already a
clear sign of accelerated eutrophication (Goldman and Armstrong, 1969). The senior author
believes that if transparency is reduced by inorganic turbidity or increased algal growth a lowering
of the lake's heat budget will occur with the possibility of freezing. This hypothesis is currently
under investigation since Tahoe does not freeze, and a winter ice cover, by promoting reduced
conditions at the mud-water interface, would probably accelerate eutrophication.
The present synoptic study is directed toward identifying the major sources of nutrients reaching
Tahoe and the patterns of eutrophication they produce in the lake. Previous studies proved to be
particularly useful approaches to the problem of identifying nutrient sources and evaluating their
effect on adjacent Tahoe waters. The investigation reported here included primary productivity of
phytoplankton and periphyton as well as examination of species composition, biomass and biotic
diversity. In addition, benthic invertebrate organisms were collected and their diversity
determined.
Synoptic studies are often made at sea where oceanographic vessels collect large numbers of
samples in transects or over broad areas. Such collections are limited in their coverage by the time
required for sampling each station and the slow speed of the vessels between stations. Further,
most of the marine synoptics cover days, weeks, or months, so that time and weather are
important variables. Some improvement is anticipated as automated sampling and aerial
reconnaissance techniques are improved, but the limitations indicated above remain serious
problems. Synoptics covering physical and chemical parameters have been made in the Great Lakes
(e.g. Ayers. et al., 1958, Anderson and Rodgers, 1963, Saunders, Trama and Bachmann, 1962).
Synoptic surveys of two New Zealand lakes for water chemistry, phytoplankton, and zooplankton
distribution were made by Fish and Chapman (1969). For primary productivity measurements,
shipboard incubation has long been used for synoptic surveys at sea, and Sorokin (1959) used the
technique in Rybinskii reservoir over a period of seven days in June. He found photosynthesis
varied in different parts of the lake as much as tenfold during this period.
Because of the photosynthetic variation from day to day, reflecting changes in the composition of
phytoplankton, nutrients, or weather, a synoptic covering the shortest possible time period will
have the greatest precision for detecting variation in fertility within the system. The synoptic
approach of 1968 reported here was probably unique in covering the entire surface layers of a
large lake in a single day without disrupting the natural light cycle and with an in situ incubation
of the primary productivity samples. This eliminated the uncertainties associated with studies that
covered longer time periods with various light conditions, or the necessity of using the unnatural
-------
light of deckboard incubation. It provided a nearly instantaneous measure of productivity, and
phytoplankton composition and biomass over the whole lake.
METHODS
The synoptic approach evolved from the first series of synoptics done in 1962 to the 1968 series
reported here. In 1962 eight stations were incubated in situ with varying incubation time from
station to station. These were corrected to a daily value on the basis of a diurnal series of samples
run at an index station. In 1967 five different stations on three different dates were sampled for
comparison of primary productivity. Incubations were done in a rotating incubator under constant
artificial light at surface water temperature. Samples for each station were taken at two depths
where the photometer measurements indicated 75% and 50% transmission of the available surface
light.
The following paragraphs describe the methods used in 1968.
Phytoplankton
In the summer of 1968 three synoptic studies were conducted: synoptic one during 16-17 July
synoptic two during 11-12 August and synoptic three during 1-2 September.
Thirty stations were selected, twenty-five of which were located along the shore around the lake.
Five stations were in the middle of the lake. In areas where it was desired to test for steep
gradients in production, three stations were located relatively close together, such as for General
Creek, Upper Truckee, Cave Rock, and Incline Creek where one station was chosen at the mouth
of the creek and one on each side.
Two boats were used to collect all the water samples within about a six hour time period. All the
water samples were collected at night, between approximately 8:30 pm and 2:30 am, and stored in
insulated boxes, so that no photosynthesis or sample warming would take place and so as not to
expose the organisms to high surface light or disrupt their diurnal rhythm. Four depths were
sampled: 0 meters, 5'meters, 10 meters, and 15 meters. Water samples were collected in 125 ml
Pyrex bottles for measurements of primary productivity using the carbon-14 method as modified
by Goldman (1963). Phytoplankton and water chemistry samples were collected at the same time.
All samples were returned to the index station located near Homewood on the west shore. This is
near our laboratory where regular in situ productivity measurements are taken on a year round
basis.
All primary productivity samples were rapidly injected with carbon-14 solution just before sunrise,
lowered into the lake, and suspended at the same depths from whence they were collected. The
bottles were left to incubate all day and were returned to the laboratory for filtration at sundown;
handling of the large number of bottles was therefore accomplished before any occurrence of
photosynthesis and after photosynthesis had stopped in all bottles. This provided an equal
incubation period for all samples so that the primary productivity measurements for all stations
were comparable.
Phytoplankton samples were collected for each station by pooling aliquots from each depth into
one sample. The samples were fixed with neutral Lugol's solution and returned to the University
-------
4
of California campus at Davis for identification, counting, and measurement of individuals for
determination of their biomass. An inverted microscope was used for this work.
The water chemistry samples for alkalinity determination were taken for each station at the 5
meter depth only, since no significant change with depth down to 15 meters was detected in Lake
Tahoe. Carbon available for photosynthesis was calculated from these samples by titration with
acid (Saunders, Trama, and Bach ma nn, 1962). We have since gone to infrared CO2 analysis which
is more rapid.
Periphyton
Seventy-nine sampling stations were chosen around the lake along the shore and samples taken
between 28 March 1968 and 18 September 1968. Periphyton samples were collected on glass
cylinders of 40 mm lengths and 24 mm O.D. Pyrex held by test tube holders at a depth of five
meters. The pieces of glass tubing were cleaned with hot acid and stored in 0.1 N Hcl prior to use.
The length of time required for the development of significant attached growth varied according to
the productivity of the habitat. A week to ten days incubation was found to be adequate at the 5
meter depth in Lake Tahoe. When the periphyton collectors were retrieved, they were stored
frozen in individual plastic snap cap vials until combusted for total carbon content. Combustion of
periphyton was achieved by placing the tubing in a .005-inch-thick platinum sleeve in an induction
furnace (LECO Series 521). This rapid method for the determination of the carbon content of
periphyton depends on the measurement of the combusted carbon in an infrared gas analyzer
(Armstrong. Goldman, and Fujita, in press).
Benthos
Thirty-nine locations were chosen, evenly distributed over the lake on a grid, plus one station in
Emerald Bay and one on a "lakemount" (discovered and described by C.R. Goldman and J. Court,
1968). Sampling was conducted between August 1.1968, and October 26, 1968 and served as the
basis for a study of sedimentation in the lake (Court, Goldman, and Hyne, manuscript). Only one
sample was taken from each station using a Shipek sampler, so that a high degree of quantitative
reliability cannot be claimed for the collections. The sediment samples collected varied in amount
from less than one liter to more than four liters, depending on the substrate, and were washed and
sieved before sorting. Lake water was used for the washing, and care was taken not to cause a
grinding of the organisms in the washing of the sand and silt. Sieving was graded and fine enough
to prevent the passage of even such small organisms as ctadocerans and copepods. Following the
washing, the samples were preserved in formalin In smarf vials. The various organisms were then
identified and enumerated.
RESULTS AND DISCUSSION
In 1967 the variation in primary productivity, was particularly high in August and September.
Comparisons were not converted to mgC/m2/day because they were incubated under constant
light not very representative of the lake conditions. Crystal Bay and the South Shore were the
highest stations in August and early September respectively (Fig. 2). With the onset of winter
storms and mixing, the variation in stations disappeared in late September.
The 1968 synoptics provided a great deal more data than the 1967 studies. For the synoptic at
-------
AB.-Ago1« Boy
SH.-Skunk Harbor
S.S.-Sou1h Shore
CBrCrystal Bay
M-LrMld-Loki
FIGURE 2 Primary productivity measurements taken at five different locations of LakeTahoe,
on three different sampling periods, at the two depths where 75% and 50% light are
transmitted.
-------
each station, primary productivity for the day, under a square meter surface, for the 15 meter
column of water, was calculated in mgC/m2/day. A contour map of primary productivity was
constructed from this data (Fig. 3). For synoptic one, the range of values was from 11.04
mgC/m2/day in a midlake station to 78.52 mgC/m2/day at the mouth of Incline Creek. For the
second synoptic, values ranged from 12.41 mgC/m2/day in a midlake station to 90.86
mgC/m2/day at Tahoe City. For the third synoptic, values ranged from 17.85 mgC/m2/day in a
midlake station to 84.07 mgC/m2/day at the mouth of the Upper Truckee River.
In 1968, during each synoptic study, primary productivity was measured at 13 different depths at
the index station, and the percentage of the primary productivity of the entire euphotic zone
represented by the primary productivity of the first 15 meters was estimated from this
comparison. On that basis, the average primary production of the euphotic zone, extended to 105
m (the average depth of the euphotic zone), was found to be 125.0 mgC/m2/day for the first
synoptic, 109.6 mgC/m2/day for the second, and 193.9 mgC/m2/day for the third.
For each station, the average of the three synoptics was calculated and a contour map of primary
productivity in Lake Tahoe was drawn based on the average of the three synoptics (Fig. 3). The
range of average values varied by fourfold, from 14.90 mgC/m2/day in a midlake station to 57.22
mgC/m2/day at Tahoe City. The spatial patterns of primary productivity appeared to be not
random but ordered. Primary productivity is often high near shore and at the mouths of creeks. It
is generally low at stations near the middle of the lake.
The total number of individual phytoplankton per ml and the total fresh weight (biomass) in
milligrams per cubic meter were calculated for each synoptic at each station. Average values for
the three synoptics at each station were used to draw isopleth lines of phytoplankton on the map
of Tahoe (Fig. 4), as was done for primary productivity. Average values of number of individual
per ml varied by 4.5 fold, from 20.58 in midlake to 77.08 at the mouth of Incline Creek and 89.33
at Tahoe City. Average of three synoptics per station for the biomass of the phytoplankton varied
by sixfold, from 19.89 mg/m3 at Rubicon Point and 20.65 in midlake to 77.71 mg/m3 at the
mouth of Incline Creek and 129.56 mg/m3 at Tahoe City.
Spatial patterns of phytoplankton, like the primary productivity, appeared ordered.
Phytoplankton concentration was high near shore and at the mouths of some creeks. It was
systemically low in the middle of the lake.
Biotic diversity per individual on the number of individuals per species of phytoplankton was
computed for each station and each synoptic. The expression for diversity used here is the
information measure of diversity used by Margalef (1957, 1965), Goldman et al. (1968) and
Goldman (1970, in press).
i=m nj nj
D = - 2 Iog2 0
-------
17 JULY
2 SEPTEMBER
Average of the 3 Synoptics
FIGURE 3 Contour map of primary productivity in the upper 15 meters of water based on
mgC/m2 /day1 measured at 30 stations for each of the 3 synoptics. The fourth map is
based on values at each station which are the average of the 3 values of the individual
synoptics. X's indicate the station locations.
-------
CD
Numbn 01 IndMduolt
Pwml
filomois
mq. m"'
Blotlc Div«r»ily Pir Individual
Bill Pv C«ll
FIGURE 4 Phytoplankton number of individuals, biomass, and diversity per individual. Isopleth
lines have been drawn using values at each station which are the average of the 3
individual synoptic values. X's indicate the station locations.
-------
Average values for the three synoptics at each station were used to draw isopleth lines of biotic
diversity in bits per individual on the map of Lake Tahoe, as was done for the other variables (Fig.
4). Average values of biotic diversity per individual varied from 2.61 bits per individual in a
midlake station to 3.59 bits per individual at Tahoe City. Spatial patterns of diversity appeared to
be not random but ordered. Higher values were largely in shallow water areas near shore and at the
mouths of some creeks where productivity was also higher. Diversity values were systematically
low at some of the deep midlake stations.
A considerable difference between shallow water and deep water productivity and phytoplankton
concentration is evident from Figures 3 and 4.
A student t test was used to determine the significance of this difference between sampling
stations in midlake and those nearer the shore. Four combinations of stations were tested.
I. The four midlake stations of the deep northern part of the lake versus all other stations.
II. The four midlake stations of the deep northern part of the lake plus the midlake deep
station offshore from the Upper Truckee River versus all other stations.
III. The four midlake stations of the deep northern part of the lake plus the three stations
alongshore off General Creek versus all other stations. General Creek on the West Shore
of the lake is a tributary of low nutrient concentration and of small stimulation to algal
growth (Goldman and Armstrong, 1969).
IV. Same as III but including one more station south of General Creek close to shore, and
the midlake station off the Upper Truckee River with the other midlake stations,
namely 8 stations, versus the other 22 stations.
Differences in primary productivity, concentration of phytoplankton individuals, concentration of
phytoplankton biomass, and diversity per individual of phytoplankton were tested. Results are
shown in Table 1. Data from all three synoptics and all thirty stations per synoptic (90 data
points) were used for each t value determination. Significant results were obtained from all four
groupings selected for primary productivity, phytoplankton number of individuals, and biomass
concentrations. This indicated that the difference between the means of the two groups of stations
selected in all four cases was sufficient to warrant the conclusion that the one group with midlake
stations was definitely different from the other. Highest significances, 0.1% and 0.5%, were
obtained for primary productivity. Grouping arrangements I and II gave the most significant
results for primary productivity and grouping IV gave the most significant results for
phytoplankton concentration. For the data on phytoplankton diversity, differences between the
means of the two particular groups of stations selected in those four cases were not sufficient to
conclude that one group was different in diversity from the other. This was probably due to the
one midlake station on the northwest of the lake where the diversity was quite high.
These results show the highly significant differences in primary productivity between the shallow
and deep water stations of the lake, as well as the difference in phytoplankton numbers and
biomass. Furthermore, they show that primary productivity and phytoplankton concentrations in
the waters along the shore off the General Creek area were not as different from midlake results as
those of the waters sampled offshore near other disturbed, drainage areas. They were more similar
-------
10
TABLE 1 fTEST RESULTS
Significance of Difference Between Sampling Stations in
Midlake and Along the Shores
Primary Productivity
I
Stations
1-25,30
versus
26-29
4.074
0.1%
II
Stations
1-25
versus
26-30
3.906
0.1%
III
Stations
1,5-25,30
versus
2,3,4,26-29
3.286
0.5%
TV
Stations
1,6-26
versus
2-5,26-30
3.163
0.5%
Phytoplankton No. Individuals
2,009
5%
2.317
2.5%
1.862
10%
2.653
Phytoplankton Biomass
1.706
10%
1.896
10%
1.851
10%
2.470
2.5%
Phytoplankton Diversity
1.240
1.477
0.388
1.418
Results from all three synoptics on a total of 30 stations were used for each T value determination.
Level of significance is for 88 degrees of freedom and is given as percent level under the t value.
-------
11
to the midlake deep water stations than other shoreline waters, since grouping IV also gave
significant results, particularly for primary productivity.
Higher diversity (Fig. 4) was evident in the shallower water areas of the lake as well as in the
vicinity of streams where productivity was higher due to the inflow of nutrients from the
watershed. This pattern of diversity in Lake Tahoe was observed not only in the average values
from the three synoptics; each synoptic cruise showed the same general pattern of diversity. The
work of Margalef (1964) suggests that the opposite pattern of species diversity should have been
found. That is, higher diversity occurring in the less enriched, less productive central waters of the
lake and less diversity in the more enriched, more productive shoreward waters of the lake. This
position is certainly borne out by our studies of a highly productive lake like Clear Lake,
California, which at times may be almost a monoculture of bluegreen algae such as
Aphanizomenon. However, in highly oligotrophic environments such as Tahoe, a eutrophicating
influence such as fertile stream water may increase the diversity in the immediate area. Similarly,
benthic forms may enter the lake directly from the streams. The relative importance of these
benthic forms to the number of individuals, the biomass of the phytoplankton and directly to the
diversity values was determined. Out of the one hundred and three different species of
phytoplankton found most commonly in Lake Tahoe, nine were most likely to be benthic forms
that became dislodged from their attachment by wave action or otherwise freed from the
substrate. These were mainly diatoms and included Cocconeis placentula, Cocconeis disculus,
Gomphonema parvulum, Gomphonema capitatum, Cymbella ventricossa, Cymbella lanceolata,
Cymbella cuspidata, and Diatoma vulgare. Some Anabaena have also been found in the benthos of
Lake Tahoe. The percentage of those species found in the total phytoplankton population in terms
of number of individuals and biomass was calculated at each station for each synoptic study. The
same percentages were also calculated from the average values of the three synoptics in number of
individuals and biomass at each station. Those results are shown in Figure 5. A comparison of
these results with a bathymetric map of the lake does not show a very close correlation between
shallow water and high percentage of benthic forms. Conversely, deep waters and low percentage
of benthic forms are not closely correlated, although the highest percentages are found in the
shallower waters. This may in part be due to the fact that shallow waters are only a small part of a
lake whose mean depth is 313 meters. The curious pattern of diversity in Crystal Bay may well
reflect the deep canyon running out from shore where benthic forms may not enter the catch so
frequently. Wind patterns and water movements might also influence the presence of benthic
forms in the phytoplankton population of the different areas of the lake, as might the nutrient
history of existing periphyton of the various areas. This is particularly true if we consider that
periphyton has only become abundant in the lake during the last ten years as the nutrient content
of littoral water has accelerated its growth. Comparing these results with the biotic diversity per
individual did not show a systematic relationship. To verify and further investigate this, the biotic
diversity per individual was calculated again for each synoptic cruise and for the average of the
three, but this time omitting the benthic species. All new values were lower than those obtained
from the entire phytoplankton population, but the overall pattern of diversity of the lake
remained essentially unchanged.
Rate of growth of periphyton varied from .094 mgC/m2/day, north of Skunk Harbor between 20
July and 18 August, to 2.228 mgC/m2/day, 70 yards offshore north of Sand Harbor between 15
July and 23 July. Periphyton growth was found to be a function of both the location and the time
of incubation period (Fig. 6).
-------
12
2 SEPTEMBER S ' AVERAGE OF THE J SYNOPTICS
SCALE
STATION LOCATION , 10* NUMBER Of INDIVIDUALS
JO* BIOMASS
FIGURE 5 Percent number of individuals and percent biomass of benthic forms of algae found
in total phytoplankton counts, for each synoptic study and on the average.
-------
A high initial loss of samples from periphyton stations resulted in considerable variation in
effective sampling effort. Later the original surface floats were all replaced with underwater floats
tended by scuba divers. As a result of loss of stations, quite a few values for the rate of growth of
periphyton were missing from the overall matrix of data which consisted of rows of stations versus
columns of incubation periods. In order to be able to determine variations of rate of growth of
periphyton between areas along the shores of the lake, three types of computations were made.
TAHOE PERIPHYTON 1968
Pinodi of incubation
O J»l» I*—July 29
Jul, I8-A00..7
Jul,20-Aiig. 18
AugJS-Stpt. I
Aoo.28-S.pt. 13
Periods of incubation
^. June 18 —July 17
June is-July 10
July 4 — July 18
0—July '*
5 -July 23
FIGURE 6 Periphyton rate of growth
-------
14
First, adjacent stations of similar productivity were pooled and averaged to give one value for that
particular area for the period of incubation (pooling of rows of the matrix). Seventeen main
stations resulted from this sample pooling. Second, periods of incubations adjacent in time and
showing similar rate of growth for most stations were pooled to give an average rate of growth for
the entire period (pooling of columns of the matrix). Third, missing values for one of the
seventeen main stations at one of the resulting five periods of incubation were interpolated from
the other values. Interpolations were made favoring rows rather than columns, namely using more
points in time than stations, rather than a large number of stations and fewer points in time. After
completing the matrix of 17 stations and five incubation periods, columns were averaged to give
one average value of rate between 28 March and 18 September 1968 (Fig. 6). These results vary
between .343 mgC/m2/day northeast of the Upper Truckee River to 1.010 mgC/m2/day south of
Tahoe City.
The average daily amount of growth of periphyton on the bottom of the entire lake was estimated
by use of a planimeter for each area between the shore and the 100-meter depth, multiplying the
rate of growth of that location by the area found, and summing up the results. The weight of
carbon of algae was estimated to be 13% of the total fresh weight of the algae. On that basis the
average total fresh weight of periphyton added daily to the bottom of the lake (by photosynthesis
on the bottom surface only) was found to be about 576.8 kilograms. This was an underestimate of
the bottom average daily growth of periphyton, since no correction was made to account for the
irregularity of rock surfaces. Fox, J. L.r Odbug, T. O. and Olson, T. A. (1969), found in western
Lake Superior that, "After forty-six days of regrowth on artificially denuded rocks in Stony Point
Bay, the growth level was approximately eighteen percent of that occurring naturally." Assuming
this result valid for Lake Tahoe would have meant that there are about 147,397 kilograms (fresh
weight) of periphyton in Lake Tahoe (about 147 metric tons). For purposes of comparison the
biomass of planktonic algae was calculated by averaging biomass per square meter for the three
synoptic dates, which resulted in an estimate of 2,180 metric tons in the upper euphotic zone of
the lake.
Thirty-three different genera representing a total of twelve different classes of animals were
identified in the benthos samples (Table 2). Two genera composed 46.6% of all identified
organisms. The dominant organism found in 30 of the 41 stations was Stygobromus (Amphipoda).
A total of 219 of these identified, which represented 27.5% of all the identified organisms and
24.5% of all organisms found. The second dominant organism found in 26 stations wasCandona
(Ostracoda). One hundred and fifty two were found which represented 19.1% of the identified
organisms.
The total number of organisms found in any one sample varied from 0 to 79 (Fig. 7). Values of
diversity per individual were computed for each station and varied from 0 to 2.94 bits per
individual (Fig. 8).
Values of diversity per individual for our data on benthos were probably more relevant than total
number of individuals per sample, due to the limited number and difficulty of obtaining samples
of comparable size.
-------
IE
DIAMETER SCALE
_J1120INOIVIDUALS
FIGURE 7 Lake Tahoe 1968 Benthos - Total Number of Individuals Per Station Per Sample
DIAMETER SCALE
,_, I BIT PtR INDIVIDUAL
FIGURE 8 Lake Tahoe 1968 Benthos Diversity Per Individual
-------
16
TABLE 2
Organisms Found in the Benthos Samples
CLASS
Protoza
Turbellaria
Nemata
Oligochaeta
Ostracoda
Copepoda
Malacostraca
Amphipoda
GENERA
Tetrameris
Phagocata
Dendrocoelopsis
Rhynchoscolex
Trilobus
Dorylaimus
lotonchus
Peloscolex
Lumbriculus
Limnodrilus
Herdea
Ilyiodrilus
Rhynchelmis
Haplotaxis
Naidium (Breviseta)
Eisentetta
Aeolosoma
(Immature)
(Newly hatched)
(Cocoons)
(Eggs)
Candona
Diaptomis (Tyrelli)
Epishura
Stygobromus
Hyalella
(Decomposed)
(Unidentified)
NO. OF
STATIONS
ORGANISMS TOTAL NO.
PRESENT ORGANISMS
20
1
1
8
8
8
19
3
4
2
5
3
1
1
1
1
3
4
9
2
26
1
1
30
9
2
2
79
3
1
25
23
20
57
26
24
19
13
8
5
2
2
1
19
20
24
2
152
37
3
219
28
2
2
% OF
TOTAL NO.
ORGANISMS
.8
83 9.3
68 7.6
222 24.9
152 16.9
40 4.5
251 28.2
Insecta (Chironomidae) Orthocladiinae * 6
Procladius 2
Chironomus (Atritibia) 1
Phaenopsectra 1
17
2
1
1
-------
17
CLASS
Insecta (Hecoptera)
Acari
Mollusca
Gastropoda
Pelecypoda
Terrestrial Arachnid
Miscellaneous
TABLE 2 (Cont'd)
Organisms Found in the Benthos Samples
GENERA
Capina
(Terrestrial Gnat)
(Midge)
(Larva)
(Diptera Adult)
(Unidentified)
Lebertia
Limnochares
Parapholyx
Gyraulus
Pisidium
Arachnida *
Eggs
Cocoons
NO. OF
STATIONS
ORGANISMS TOTAL NO.
PRESENT ORGANISMS
% OF
TOTAL NO.
ORGANISMS
2
1
1
1
2
1
9
1
2
1
6
1
4.4
1
18
3
33 3.7
3 .3
7 .8
.1
21 2.3
Total No. of Organisms Found 892
*Not identified to Genus
CONCLUSIONS
The synoptic approach is very useful in studying cultural eutrophication, as it enables the
investigator to accurately locate sources of nutrients before the entire lake has undergone change.
Lake Tahoe showed several areas of increased fertility. These were at the South Shore under the
influence of the Truckee River drainage and high resident population, in Crystal Bay where Incline
Creek and Third Creek drained highly disturbed land, and near the outflow of the lake where there
were both a high resident population and fairly extensive areas of shallow water.
Despite the lake's great volume for dilution of the annual inflow, local nutrient sources are altering
the productivity pattern around the lake. A spring bloom of algae, which actually turned a path of
Tahoe's traditionally blue water green, has already been observed at the south shore and
periphyton growth has become luxuriant around the entire lake margin during the last decade. In
August 1969, following extensive land disturbance associated with construction of a golf course
and a subdivision, Third Creek was found to be extremely stimulating to growth of Tahoe algae. In
the experiment summarized in Figure 9, 10% of this stream water added to Tahoe's natural
phytoplankton population stimulated photosynthesis by over 600%. When compared with
-------
18
controls. Incline Creek and the Upper Truckee River were also notable lake fertilizers in contrast
to relatively undisturbed General Creek. This is not surprising since the Incline Creek drainage has
been recently subdivided for homes and the Upper Truckee River and some of its tributaries have
also suffered considerable disturbance. Further, this drainage was used for some time as a land
disposal site for treated sewage. The sewage is now being exported out of the basin, but the land
disposal site still provides a drainage high in nitrogen.
Although occassional high periphyton values are encountered near tributaries, there appears to be
less correlation with tributaries than was found for phytoplankton productivity and biomass. In
general, distribution of periphyton is fairly uniform around the lake. This probably reflects the
steady movement of water over the littoral zone of the lake which distributes the nutrients rather
uniformly to these sessile forms.
The results of our study of the possible relationship between diversity and benthic forms can only
be considered as a preliminary investigation since our list included only those species that were
most likely to be found in the periphyton of Lake Tahoe. Further, more detailed observation of
the lake's periphyton species composition at the time of the study would have been a valuable
addition.
From the above results we can say that benthic forms of algae contribute overall to increase the
diversity per individual of the Lake Tahoe phytoplankton communities, but that they do not
appear to be the major determinants of aerial variation in diversity.
740 -]
700-
t 660-
| 620-
« 580-
f 540-
| 500-
« 460-
« 420-
u
£ 380-
S« 340-
i 300-
2 260-
| 220-
S 180
"5 140
o>
< 100-
RD-
' 0-eek
lociifift
Village
LAKE TAHOE
August 12, 1969
Incline
Creek
Incline
Village
Mfcrtr
Truckee
River
Shore Cr*efc
Unpolluted
• '< • Cpntol -
10% STREAM WATER ADDED TO LAKE WATER
FIGURE 9 Fertilization of Lake Tahoe water from a 10% addition of four of its tributaries.
Algal growth rate was measured with the natural Tahoe population by radioactive
carbon uptake. Highest biostimulation was measured in Third Creek following
disturbance of its watershed. General Creek, which drains an area of undisturbed
land, is used as control.
-------
19
The results of the benthic survey are the most difficult to interpret, for there does not appear to
be a systematic pattern to either the distribution of numbers of organisms or the distribution of
biotic diversity. This condition is especially evident when compared to the synoptic surveys of
primary productivity where the primary productivity rate increased sharply near the mouths of
some of the creeks. It is quite possible that the lack of pattern in the benthos data may be a result
of the lack of replicated samples and the small and variable sample sizes. However, the abundance
and diversity of the benthic organisms may not be functions of the same environmental property
as the abundance, productivity and diversity of the phytoplankton. Algae are clearly associated
with the mouths of creeks and their productivity has been shown to be stimulated by additions of
creek water (Goldman and Armstrong, 1969). The environmental factors that affect the benthos
have not been determined, and a lack of pattern in benthic numbers and diversity may be an
indication that nutrient enrichment is not a primary factor limiting the abundance of various
species of benthos.
Tahoe is unquestionably changing. The synoptic approach provides a nearly instantaneous
evaluation of conditions as they exist on a given day. Close monitoring of wind and currents could
add further to our understanding of water transport in the lake as it affects the presence of more
fertile areas. In general, the primary producers must be viewed as a more sensitive indication of
increased fertility than chemical parameters, since any additional nutrients appear to move rapidly
into the phytoplankton. We characteristically find little or no measurable change in water
chemistry (unpublished data) while phytoplankton photosynthesis is showing a very significant
change. If the quality of our lakes is to be conserved, synoptic studies can be a great aid in locating
and defining the influences of nutrient sources. When combined with bioassays, the relative
stimulation of the various sources can be quantified and corrective measures taken before
accelerated eutrophication destroys the quality of the environment.
ACKNOWLEDGMENTS
The authors wish to acknowledge the valuable field assistance of James Court, Gordon Godshalk,
Richard Armstrong, and Peter Richerson. Frank Sanders sorted and identified the benthic
organisms and Anne Sands handled the phytoplankton taxonomy and counting. Denne Bertrand
and Noel Williams gave valuable assistance in data reduction. Elisabeth Stull provided valuable aid
in preparation of the final manuscript. The work was supported by FWPCA grant (now FWQA)
No. DBU 16010 to the senior author.
REFERENCES
Abrahamsson, S. A. A. and Goldman, C. R. (1970) The distribution, density, and production of
the crayfish, Pacifastacus leniusculus Dana in Lake Tahoe, California-Nevada, Oikos 21:1-9.
Anderson, D. V. and Rodgers, G. K. (1963) A synoptic survey of Lake Superior. Great Lakes Res.
Div., Univ. Mich. Pub. No. 10:79-89.
Armstrong, R., Goldman, C. R. and Fujita, D. K. (1970) A rapid method for the estimation of the
carbon content of seston and periphyton. Limnol. Oceanogr., In Press.
-------
20
Ayers, J. C., Chandler, D. C., Lauff. G. H., Powers, C. F. and Hensen, E. B. (1958) Currents and
water masses of Lake Michigan. Great Lakes Research Institute, Univ. Mich. Pub. No.
3:1-169.
Court, J. E., Goldman, C. R. and Hyne, N. J. (1970) Surface sediment in Lake Tahoe,
California-Nevada. (Submitted manuscript).
Fish, G. R. and Chapman, M. A. (1969) Synoptic surveys of lakes Rotorua and Rotaih. N.Z. J.
Mar. Freshwat. Res. 3:571-84.
Fox, J. L., Odlaug, T. O. and Olson, T. A. (1969) The ecology of periphyton in Western Lake
Superior. Water Resources Research Center, Univ. Minn., Bull. No. 14:1-127.
Frantz, T. C. and Cordone, A. I. (1967) Observations on deepwater plants in Lake Tahoe,
California and Nevada. Ecology 48:709-714.
Goldman, C. R. (1963) The measurement of primary productivity and limiting factors in
freshwater with Carbon-14. In: M.S. Doty (ed.), Proc. Conf. Primary Productivity
Measurement, Marine, and Freshwater, U.S. Atomic Energy Commission, Tl0-7633:103-113.
Goldman, C. R. (1964) Primary productivity and micro-nutrient limiting factors in some North
American and New Zealand lakes. Verb. Internal. Verein. Limnol., 15:365-374.
Goldman, C. R. (1967) Integration of field and laboratory experiments in productivity studies. In:
George H. Lauff, (ed.). Estuaries. Publ. No. 83, AAAS, Wash., D.C., p. 346-352.
Goldman, C. R. (1968) Aquatic primary productivity. American Zoologist, 8:31-42.
Goldman, C. R. (1970) Photosynthetic efficiency and diversity of a natural phytoplankton
population in Castle Lake. Proc. Trebon IBP/PP Symposium. In Press.
Goldman, C. R. and Armstrong R. (1969) Primary productivity studies in Lake Tahoe, California.
Verh. Internal. Verein. Limnol., 17:49-71.
Goldman, C. R. and Carter, R. C. (1965) An investigation by rapid C-14 bioassay of factors
affecting the cultural eutrophication of Lake Tahoe, California-Nevada. J. Water Poll. Control
Fed., 37:1044-1059.
Goldman, C. R. and Court, J. (1968) Limnological studies of Lake Tahoe. In: Geologic Studies.
Guidebook of Geol. Soc. Sacramento, 60-66.
Goldman, C. R., Gerletti, M., Javornicky, P. Melchiorri-Santolini, U. and de Amezaga, E. (1968)
Primary productivity, bacteria, phyto and zooplankton in Lake Maggiore: Correlations and
relationships with ecological factors. Mem. 1st. Ital. Idrobiol., 23:49-127.
Goldman, C. R., Tunzi, M. and Armstrong, R. (1969) Carbon-14 uptake as a sensitive measure of
the growth of algal cultures. FWPCA Symposium, Berkeley, June 1969, 158-170.
-------
21
Hutchinson, G. E. (1937) A contribution to the limnology of arid regions primarily founded on
observations made in the Lahontan Basin. Trans. Conn. Acad. Arts. Set. 33:47-132.
Juday, C. (1907) Notes on Lake Tahoe. its trout and trout fishing. Bull. U.S. Bur. Fish.,
26:133-146.
Kemmerer, G., Bovard, J. F. and Boorman, W. R. (1923) Northwestern lakes of the United States.
Bull. U.S. Bur. Fish., 39:51-140.
LeConte, J. (1883a) Physical studies of Lake Tahoe -1. The Overland Monthly,1:506:516.
LeConte, J. (1883b> Physical studies of Lake Tahoe - II. The Overland Monthly, 1:595-612.
LeConte, J. (1884) Physical studies of Lake Tahoe - III. The Overland Monthly.3:41 -46.
Margalef, R. (1957) La teria de la informacion en ecologia. Mem. R. Acad. Cie. y Artes de
Barcelona, 32:373-449.
Margalef, R. (1965) Ecological correlations and the relationship between primary productivity and
community structure. Mem. 1st. ItaL Idrobiol., 18 Suppl.:355-364.
Saunders, G. W., Trama, F. B. and Bachmann, R. W. (1962) Evaluation of a modified C14
technique for estimation of photosynthesis in large lakes. Great Lakes Res. Div. Univ. Mich.
Pub. 8:1-61.
Sorokin, Y. I. (1959) Determination of the photosynthetic productivity of phytoplankton in
water using C14. Fiziol. Rast. (U.S.S.R.), 6(1):125-133.
-------
THE SOUTH BASIN OF LAKE WINNIPEG
AN ASSESSMENT OF POLLUTION
Jo-Anne M. E. Crowe
INTRODUCTION
During their brief geologic life, lakes are subjected to a variety of physical and biological processes
which result in their ultimate extinction (Powers and Robertson, 1936; Robertson and Alley,
1966). This is a natural, slow process termed eutrophication. Eutrophication accelerated or
augmented by man's activities may be termed pollution.
Classical studies on lake eutrophication include those by Minder (1926), Hasler (1947) and
Edmondson, Anderson and Peterson (1956). The most intensive studies on the North American
continent have involved the Great Lakes and particularly Lake Erie (Wright, 1933; Brown, 1953;
Powers et al., 1959; Beeton, I960, 1961, 1963 and 1966; Matheson, 1962; Wood, 1963; Carr and
Hiltunen, 1965; Matheson, 1965; Powers and Robertson, 1966; Arnold, 1969).
Lake Winnipeg possesses many features which render it comparable with Lake Erie. Both are large
Pleistocene lakes of recent origin. In size Lake Erie ranks 13th and Lake Winnipeg 14th among the
freshwater lakes of the world. Both are adjacent to major population centers. Both are employed
as transportation systems to varying degrees. The commercial fishery of both lakes provides a
major source of food and revenue. Finally, both lakes have served as recreational areas.
In contrast to the abundance of published literature dealing with Lake Erie, little material dealing
with general limnology is available for Lake Winnipeg. Bajkov (1930) studied the chemical
composition of Lake Winnipeg waters and the benthic composition. Studies on the mayflies of
Lake Winnipeg were published by Neave (1932). Slack (1967) studied the benthos of certain lakes
in Manitoba whose basins were located astride Precambrian and Paleozoic bedrocks. Einarsson and
Lowe (1968) studied seiche cycles on Lake Winnipeg and Anon., (1969a) reviewed the problems
associated with fluctuating lake levels. An assessment of the possible effects of pollution on the
benthos was presented by Cover (1966).
It is worthy to note that Butler (1949) said, "Pollution of waters inhabited by fish is not yet a
major problem in Manitoba." It is our contention that this problem has arisen in Lake Winnipeg
and that the lake is undergoing a eutrophication cycle similar to that of Lake Erie. The main
evidence for increased eutrophication due to pollution is evident in the changes which have
occurred in the benthic fauna since 1930.
DESCRIPTION OF THE AREA
Morphology and Morphometry
Lake Winnipeg lies between 50° 22' and 53° 38' north Latitude and between 96° 11' and 99° 09'
west Longitude in the province of Manitoba (Fig. 1). The surface area is approximately 9,230
square miles (23,905.7 km2). Thus it ranks as the third largest wholly Canadian lake and the
largest in the province of Manitoba.
22
-------
23
FIGURE 1 Province of Manitoba, showing major lakes, rivers, cities and geologic zones
The lake is oriented in a northwest-southeast direction and spans a distance of 253 miles. It is
divided naturally into three areas, a north basin with a maximum width of 70 miles and a mean
depth of 49 feet (15m), a channel area with a maximum width of twelve miles and a maximum
depth of 60 feet (18.2m), and a south basin with a maximum width of 27.5 miles and a maximum
depth of 40 feet (12.2m).
Geology
Due to its geologic history and structure, Lake Winnipeg has unique qualities. First, it is a glacial
lake of recent origin having assumed its present form approximately 5,000 years ago (Elson,
1958). At present, with neighboring Lake Manitoba and Lake Winnipegosis, Lake Winnipeg
remains as the largest remnant of glacial Lake Agassiz (Fig. 1). Second, the contact line between
Precambrian and Paleozoic bedrock lies within the basin of Lake Winnipeg adjacent to the east
shore (Fig. 1). The latter situation insures that there will be variability both in the composition of
the influent waters and the fauna. These aspects have been investigated in this and similar lakes
(Rawson, 1953 and 1960; Oliver, 1960; Fedoruk, 1964; Larkin, 1964; Slack, 1965 and 1967).
-------
24
Watershed and Drainage
The drainage area for Lake Winnipeg is 370,000 square miles (958,300 km2) exfending through
portions of four provinces and three states. The major influent rivers are the Saskatchewan from
the west and the Winnipeg from the east {Fig. 1). The former contributes 39% of the total drainage
receipts and the latter 32% (Anon., 1969a). From the south, the Red River contributes 6% of the
total inflow while the remaining 23% is provided by numerous streams and rivers arising in the
Precambrian area to the east. Major drainage from the west is scanty and consists primarily of the
Saskatchewan River and the Dauphin River (Fig. 1). The lake is drained by the Nelson River to the
north into Hudson Bay (Fig. 1). In addition to the major influent sources, considerable water
moves between the lake basins as seiches (Einarsson and Lowe, 1968).
Sources of Pollution
There are three potential sources of pollution for the south basin of Lake Winnipeg. The first is
run-off from agricultural land to the south and west and is undoubtedly a major contributor
(Personal communication. Dr. G. Brunskill, Freshwater Institute, Fisheries Research Board of
Canada, Winnipeg).
The second is the city of Winnipeg which is located at the junction of the Red and Assiniboine
Rivers approximately 35 miles south of Lake Winnipeg (Fig. 1). Urban population has risen from
293,300 in 1931 to 515,661 in 1969 (Personal communication, Mr. A. S. Bready, Metro
Information, Winnipeg). In the 1930's, all domestic sewage and industrial effluents were
discharged untreated in the rivers. This situation has gradually been rectified and at present most
municipal and industrial wastes receive treatment in activated sludge plants or aerated lagoons. The
total capacity of these systems is 114 million gallons per day. By 1972, all wastes will be treated to
effect a 90% removal of B.O.D. and suspended solids (Personal communication, Mr. A. Penman,
Waterworks and Waste Disposal Division, Metropolitan Corporation of Greater Winnipeg).
The third potential pollution source is Abitibi Manitoba Paper Limited on the Winnipeg River
approximately 6 miles east of Traverse Bay (Fig. 1). The sole product is newsprint produced by a
sulfite process. Water consumption in the plant is 11,100,000 Imperial gallons/day. Treatment
facilities serve to remove bark and fibre. All effluents from the plant, except domestic sewage, are
discharged into the Winnipeg River. B.O.D. values of 7.9 ppm and dissolved oxygen values of 6.7
ppm have been recorded in the river adjacent to the discharge site. Downstream the B.O.D. was 3.1
ppm and dissolved oxygen 10.0 ppm (Personal communication, Mr. D. D. Munro, Abitibi
Manitoba Paper Limited, Pine Falls).
METHODS AND MATERIALS
Sampling Procedures
A four-mile grid pattern was adopted for sampling in the south basin of Lake Winnipeg (Fig. 2).
Sampling was performed twice each year; in March during the period of winter stagnation and in
September, the autumnal overturn. The scope of the sampling schedule from March, 1962 to
September, 1969 is summarized in Table 1. Winter sampling was conducted by bombardier and
autumn sampling by the Manitoba Fisheries Branch research vessel. The locations of transects were
-------
25
determined by shore landmarks and compass bearings (Fig. 2). Station positions were located by
odometer in the bombardier and by a combination of time and engine speed on the research vessel,
Limnological data routinely collected on location included water depth, surface and bottom water
temperatures, dissolved oxygen, alkalinity, free carbon dioxide and pH. Turbidity was measured
by Seichi disc to the nearest %-foot in the ice-free periods. Ice thickness was recorded in the March
periods.
Month
March
September
March
September
March
September
March
September
March
September
March
September
March
September
March
September
TABLET
Lake Winnipeg Sampling Schedule. March 1962 to September 1969
Year Number of Stations Grid Lines Sampled
1962
1962
1963
1963
1964
1964
1965
1965
1966
1966
1967
1967
1968
1968
1969
1969
70
63
22
63
40
73
39
66
20
65
16
34
20
33
20
34
AtoN
AtoG
A, F and K
AtoN
AtoP
AtoN
AtoF
AtoN
AtoD
AtoN
AtoD
AtoN
AtoD
AtoN
AtoD
AtoN
Water depth was measured to the nearest Vi-foot by a graduated sounding tine. These were
confirmed by a model D-11 Bendix depth recorder. Prior to 1967, water temperatures were
measured by a brass-encased thermometer in degrees Fahrenheit. Subsequent to this, hydrographic
thermometer has been employed. Alkalinity and free carbon dioxide values were determined by
standard titration methods (Anon., 1965). pH was determined by comparator using phenol red.
-------
26
0 Hentey Rittr
Traverse Boy
12345 67
FIGURE 2 Four-mile grid sampling pattern. Lake Winnipeg south basin
pH range 6.8 to 8.4, as indicator.
Two-litre bottom water samples were collected by 1,200 cc Kemmerer sampler from
predetermined stations during each period. Analyses for a maximum of 20 factors were performed
by the Provincial Environmental Health Laboratory in Winnipeg
During all periods except September 1964 and September 1967, a standard 9-inch Ekman dredge
was employed to collect benthos, fn September 1964 a Petersen dredge was employed, and in
September 1967 a Ponar dredge was employed. The Petersen dredge sampled 0.071 m2 and the
Ponar 0.0518 m2, as compared with 0.0523 m2 sampled by the Ekman. Triplicate benthic samples
were routinely taken at each station except in March 1962, September 1964, and September 1968.
Single dredge hauls were taken in March 1962. Duplicate samples were taken in September 1968
and September 1964, with the exception of four stations in the latter year. At each of these
stations six samples were obtained.
Prior to 1967, benthic samples were individually retained in 50-pound capacity polyethylene bags.
Whenever feasible, the samples were concentrated by washing through a galvanized bucket
equipped with bottom and side screens constructed of U.S. No. 30 mesh (Fremling, 1961). The
benthos and remaining debris were placed in four ounce jars containing 10% formalin as
preservative. Samples not concentrated in the field were frozen and shipped to Winnipeg where the
process was carried out in the Fisheries Laboratory.
-------
27
Since 1967, an 18-inch diameter hoop net constructed of Nitex nylon has been used to
concentrate the samples. The nylon employed has 62 meshes/inch with a mesh opening of 0.0098
inches (0.2450 mm) and is comparable to a No. 50 U.S. screen.
The use of Rose Bengal stain and preservative in the samples after initial concentration has aided in
sorting and enumeration of benthos (Mason and Yevich, 1967). This has proven to be more
effective than flotation techniques, particularly for nematodes (Ladell, 1936; Lyman, 1943;
Birkett, 1957; Anderson, 1959).
Laboratory Procedures
In the laboratory, benthic samples were further concentrated by washing through a No. 50 U.S.
bronze screen. The remaining were placed in an enamel pan and scanned by a Luxo light with
magnifier to facilitate separation of benthos into taxonomic groups and enumeration. Members of
the orders Gastropoda, Ephemeroptera, Amphipoda, Trichoptera, Conchostraca and the families
Sphaeriidae and Unionidae were examined with a binocular microscope for taxonomic purposes.
Whole mounts were made of the Nematoda, Chironomidae and Oligochaeta using CMC-10
mounting medium.1 The heads of the chironomid larvae were severed from the bodies at the
thoracic suture and mounted ventral side uppermost. Corresponding bodies were mounted in a
lateral position on the same slide (Curry, 1962).
Specialized dissecting techniques were required to display the oligochaete reproductive systems for
species identification (Brinkhurst and Cook, 1966). Modifications were made replacing Euparal or
Canada balsam with Amman's fluid, a temporary mounting medium (Personal communication,
Mrs. M. Moore (Simmons), Zoology Department, University of Toronto).
Where possible, identification was made to genus. Species identification of chironomids and
oligochaetes was required. Taxonomic references were as follows: Chironomidae (Johannsen,
1937; Townes. 1945; Johannsen et a!., 1952; Robach, 1957; Curry, 1954, 1958 and 1962; Beck
and Beck, 1966; Mason, 1968); Ephemeroptera (Neave, 1932; Leonard and Leonard, 1962);
Oligochaeta (Brinkhurst, 1963 and 1965; Brinkhurst and Cook, 1966); Gastropoda, Sphaeriidae,
Hirudinea, Nematoda, Amphipoda, Conchostraca, Chaoborinae and Trichoptera (Ward and
Whipple, 1959).
RESULTS
While other lakes possessing similar morphometric characteristics and geologic structure are termed
oligotrophic or at best mesotrophic. Lake Winnipeg exemplifies a eutrophic state (Northcote and
Larkin, 1963). Evidence for this is provided by the chemical composition of its water, benthic
standing crop, benthic composition and commercial fish production.
1 Mention of commercial sources does not constitute endorsement by Manitoba Department
of Mines and Natural Resources.
-------
28
Physical and Chemical Data
Mean values and ranges for chemical data compiled from March 1962 to September 1969 are
presented in Table 2. Detailed results for each sampling period are provided in Appendix 1.
South-north gradients for all factors except pH have been noted. Values recorded at stations on
the A grid line are four times greater than the mean for that period and as high as 30 times greater
than the minimum values recorded (Appendix 2; Fig. 2). Similar west-east gradients occur for
factors such as calcium, magnesium sodium, total hardness, bicarbonate, carbonate, chloride,
sulphate and total solids. Individual values along grid 1 are twice as large as those along grid lines
adjacent to the east shore (Fig. 2). For the period March 1962 to September 1969, maximum
values for most chemical components were recorded during 1966 and 1967 (Appendix 1).
Dissolved oxygen values normally represent saturations of 80% or greater during all periods (Table
2; Appendix 2). B.O.D. values are consistently lower than dissolved oxygen values (Table 2). No
decrease in oxygen concentration occurs as water depth increases. No permanent thermocline
forms and surface water temperatures are normally less than 2° F higher than those at the bottom.
Bottom deposits over the portion of the south basin under study consist of a sand-silt-clay mixture
in the following mean proportions: sand - 3%, silt - 31% and clay - 66% (Personal communication,
Mr. W. Michalyna, Canada Department of Agriculture, Winnipeg). Highest proportions of sand
were found at stations E-1 and N-5 (Fig. 2). The highest proportion of silt was found at station
A-3 and clay at stations F-3, G-3 and H-3 (Fig. 2). A definite south-north gradient could be
detected with respect to the silt component. Values for silt on the A grid line are from two to
three times the mean level and values of those on grid lines L, M and N (Fig. 2).
Benthos
Values for benthic standing crop are based on 102, 105 and 111 samples taken in the south basin
of Lake Winnipeg during June, July and August, 1967, respectively. The three dry weights,
including mollusc shell, were 88.18 kg/ha, 56.7 kg/ha and 52.24 kg/ha. Mean standing crop was
65.7 ± 3.93 kg/ha.
Between March 1962 and September 1969, the major benthic groups were Chironomidae 40%,
Gastropoda 15%, Oligochaeta 14%, Ephemeroptera 13% and Sphaeriidae 4% (Table 3). The group
identified as "others" included representatives from six orders and two families which, combined,
comprised a lower percentage than any single group (Table 3). From March 1968, due to increased
numbers, it was necessary to consider the Nematoda as a separate group.
Members of four subfamilies were among the Chironomidae identified (Appendix 2). A total of 41
genera and species were identified; 24 (59%) belonged to the subfamily Chironominae, 10 (24%)
to the subfamily Tanypodinae, 6 (14%) to the subfamily Orthocladiinae and one (3%) to the
subfamily Diamesinae.
Two species of mayfly nymphs, Hexagenia limbata occulta (Walker) and Hexagenia rigida
(McDunnough) were identified (Neave, 1932). 94% of the nymphs were H. limbata and 6% were
H. rigida. The ratio of females to males was 4:1.
The majority of Oligochaeta identified belonged to the family Tubificidae. The species identified,
-------
29
TABLE 2
Ranges in Values for Chemical Data (mg/l), Lake Winnipeg South Basin,
1930 and March 1962 to September 1969 (Mean Values in Brackets).
1930 (Bajkov) March 1962 to September 1969
True Color P* Co Units
Apparent color R Co Units
Turbidity
PH
Calcium
Magnesium
Sodium
Iron
Total hardness
Phosphate (ortho)
Alkalinity
Bicarbonate
Carbonate
Chloride
Carbon dioxide
Dissolved oxygen
B. O. D.
Sulphate
Spec cond jumhos
Total solids
Suspended solids
Dissolved solids
in order of occurrence, were Limnodrilus hoffmeisteri (Cl a parade), Tubifex tubifex (Muller),
Tubifex kessleri (Hrabe) and Limnodrilus udekekiamus (Claparede). A few specimens, tentatively
identifed as Peloscolex multisetosus (Smith) were obtained. The other family represented was
Lumbriculidae which constituted less than one percent of all oligochaetes. The species identified
was Limbriculus variegatus (Muller). During some sampling periods, immature forms comprised
50% of all oligochaetes taken. Identification, based on number, size and shape of setae, permitted
generic recognition.
-
-
-
'
13.7 to 62.8 (25.2)
2.75 to 33.7 (12.1)
-
0.14 to 1.6 (0.79)
-
-
-
-
28.2 to 128.1 (56.4)
0 to 27.5 (8.4)
-
-
-
3.9 to 118.5 (32.4)
-
100 to 480 (189.3)
-
-
15
20
3
6,60
14.4
0.28
2.0
0.1
41
0
1.8
40
0
1.0
0.1
0
0.4
3
110
38
1.5
74
to 100
to 70
to 130
to 8.99
to 60.0
to 21.9
to 18.8
to 3.9
to 224
to 1.3
to 226
to 176
to 4.8
to 19.8
to 44
to 17.7
to 8.4
to 60
to 580
to 566
to 62
to 374
(30)
(45.5)
(21.7)
(7.79)
(28.3)
(10.5)
(9.4)
(0.79)
(111)
(0.21)
(96.2)
(89.1)
(1.2)
(7.6)
(6.4)
(11.0)
(2.86)
(31.8)
(259)
(173)
(12.8)
(177)
-------
TABLE 3
Composition of Benthos, Lake Winnipeg South Basin, March 1962 to September 1969
(numbers in brackets represent percents of total numbers for that period).
Number of
ivionin
March
Sept.
March
Sept.
March
Sept.2
March
Sept.
March
Sept.
March
Sept.3
March
Sept.
March
Sept.
Mean
1 -
2 =
3 =
4 =
i ear
1962
1962
1963
1963
1964
1964
1965
1965
1966
1966
1967
1967
1968
1968
1969
1969
samples
68
147
42
186
96
150
120
202
60
189
48
99
60
64
57
105
umronomidae .
1053 (61)
835 (48)
576(78)
1057 (15)
1085 (39)
353
207 (10)
3498(43)
396(57)
5889 (74)
620 (66)
748 (45)
1409 (30)
622(13)
1709(17)
1488 (30)
(40)
Includes Trichoptera, Hirudinea,
Peter sen dredge
employed
Cipnemeroptei
288 (17)
184(11)
44(6)
1915(28)
417(15)
713(21)
449(22)
573(7)
79 (11)
779 (10)
28(3)
339(21)
39 (<1)
154(3)
45 «1)
295 (2)
(13)
Anuphipoda,
•a Oligochaeta
263(15)
397 (23)
2(<1)
1153 (17)
1 (<1)
270 (8)
0
786(10)
65(9)
811(10)
176(19)
189(12)
375(8)
1359(27)
2852(29)
1524(31)
(14)
Gastropoda Sphaeriidae Mematoda Others1
29
K<
2(<
1598
1014
1473
1151
2182
(2)
Zl)
CD
(23)
(37)
(44)
(58)
(27)
143(20)
4 (<
:i)
12(1)
133
114
212
518
(8)
(3)
(4)
(5)
522 (11)
(15)
52(3)
208(12)
18(2)
986(14)
213 (9)
523(16)
162(8)
930 (12)
0
329 (4)
89(10)
232 (14)
0
402 (8)
673 (6)
709 (14)
(4)
Unionidae, Chaoborinae, Conchostraca
34(2)
109 (6)
92(12)
112(3)
21 (<1)
26 (<1)
15(2)
109(1)
7(3)
107(2)
15(1)
5 (<1)
2473(54) 190(11)
2196(44) 24 (<1)
4208(42) 19(1)
339(7) 106(5)
(37)4 55
and Nematoda, 1962
Numbers
1719
1734
734
6821
2777
3358
1984
8078
690
7919
940
1646
4600
4969
9924
4983
3312
to 1967.
Numbers
/m*
483
225
332
701
552
315
315
765
224
801
374
318
1464
1468
3225
875
776
Ponar dredge employed
based
on percents from 4 years
-------
31
The Gastropoda were composed of five families, Physidae, Lymnaeidae, Planorbidae, Valvatidae
and Butimidae. The genera identified were Physa (Draparnaud), Aplexa (Fleming), Lymnaea
(Lamarck), Gyraulus (Charpentier), Helisoma (Swainson), Valvata Multer) and Amnicola (Gould
and Haldeman). The genera Amnicola, Physa and Valvata together composed 90% of the
gastropods.
Three genera, Sphaerium (Scopoli), Musculium (Link) and Pisidium (Pfeiffer) had equal
representation among the Sphaeriidae. All Trichoptera were tentatively identified as the genus
Limnephilus (Leach) and all Nematoda as the genus Dorylaimus (Dujardin). Members of the family
Unionidae identified were Anodontagrandis (Say), Anodontasp. (Lamarck),Lampsilis siliquoidea
(Barnes), Actinonais carinata (Barnes) and Anodontoides ferusaciana (Lea). The Hirudinea
included the species Placobdella rugosa (Verrill), Helobdella stagnalis (Linnaeus), Macrobdella
decora (Say) and Erpobdella (Blainville).
All Conchostraca were identified as Caenestheriella prob. mexicana (Glaus) and the Amphipoda as
Gammarus. Occasionally, amphipods resembling Pontoporeia were found (September, 1964).
Their exact identity is in doubt. It is feasible that these were errant migrants from the north basin
displaced by seiche action.
Fish Production
On April 20, 1970, a directive was issued by the Minister of Mines and Natural Resources ordering
closure of the summer commercial operations on Lake Winnipeg due to mercury contamination.
Mercury levels in excess of the accepted value of 0.5 ppm were found in walleye, northern pike,
sauger and perch (Personal communication, Dr. G. Bligh, Freshwater Institute, Winnipeg).
Chloralkali plants outside of Manitoba are believed to be the cause for this contamination.
The commercial fishery of Lake Winnipeg is the largest in the province of Manitoba in terms of
production, species, value and manpower (Anon., 1969b). Mean annual production from 1931 to
1969 was 13,672,371 pounds (Fig. 3). The major species by catch and value are sauger,
Stizostedian canadense (Smith), walleye, Stizostedion vitreum,wh\tef\sh,Coregonusclupeaformis
(Mitchill) and tulibee or cisco, Coregonus artedii complex. Average annual production for these
species was 4,494,735, 3,367,026, 2,311,534 and 1,549,578 pounds respectively. Other species
taken are white perch, Roccus americanus (Gmelin), bullheads, Ictalurus nebulosus (Lesueur),
carp, Cyprinus carpio (Linnaeus), catfish, Ictalurus punctatus (Rafinesque), burbot or ling. Lota
lota (Linnaeus), pike, Esox lucius (Linnaeus), goldeye, Hiodon alosoides (Rafinesque), perch,
Perca fluviatilis (Linnaeus), sucker, Catostomus commersoni (Lacepede), sturgeon, Acipenser
fulvescens {Rafinesque) and trout, primarily rainbow, Salmo gairdneri (Richardson). Since 1961,
there has been a marked decrease in fishing success (Fig. 3).
Hydrographic Data
Mean monthly levels for Lake Winnipeg, 1930 to 1969, are shown in Figure 4. Since 1961, just
prior to the initiation of the present monitoring program, lake levels were below the 56 year
average. Since 1961, there was an upward trend in water levels which culminated in the record
level in 1966. At the end of July 1966 the lake was 717.6 feet above M.S.L. This was more than
three feet higher than the normal for that time of year (Anon., 1969a). The rise continued and on
September 1, 1966 strong north winds raised the level to 721 feet in the south basin (Anon.,
1969a).
-------
32
2I-O
i
o 180
Q.
| ISO
'«•»
1 120
\
&0
'.
-
-
•
"f
TV
1 :i
* 1
;;
ll
;;
'I i
\ \ t.
•;•'
• '
;!
;i
P:
! U i
• ii i
;M
;:ii i
i * i
> • •
','.':
i;
j;
•,;\
'•i'
;j
i:
;j
iii
J '
|
i;i
X
!!
ii;
iii
:;
•H
i i
;:;]
iii
•Ii
ill
II* •
i'iii
i i:
• i
r' •'
^ 1 1
ill
iiij
; ;
fTTT
,';i':
ill
iii
P— ,
iii
I:'::
i;ii
i;
;*
i;i
'*•'•'
i'ii
i
if;
ijil
—
\
•.'.'.
I '*'
ii;i
:;:
' i ,
iii
r-^«
;'•'
:;:,
\ll\
•''','..' 1 1 ' i • •' "* '• • ' ' ' •
•'"•••''ji.'j ; i. .' • <«' >• ^^ •"'"•
I
rears
FIGURE 3 Annual commercial fish production, Lake Winnipeg 1962-1969
OBSERVATIONS
On the North American continent. Lake Erie is an authenticated example of a lake whose
environment is deteriorating as a result of pollution (Beeton, 1963; Arnold, 1969). Three indices
of pollution - chemical and physical data, benthic composition and densities, and commercial fish
production - will be examined in Lake Winnipeg and the results compared with those for Lake
Erie.
Physical and Chemical Data
Major increases in calcium, magnesium, combined potassium and sodium, sulphate, chlorides and
total solids in Lake Erie between 1906 and 1958 are well documented (Lewis, 1906; Dole, 1909;
Leverin, 1947; Thomas, 1954; Fish, 1960; Beeton, 1961 and 1963). Turbidity measurements have
been conducted and increases noted (Chandler, 1940, 1942a and b, 1944; Powers et al., 1959;
Fish, 1960; Kramer, 1961). Evidence supplied by dissolved oxygen determinations supports the
hypothesis of severe oxygen depletion over large portions of the Lake Erie basin (Britt, 1955a;
Wright, 1933; Powers et al., 1959; Fish, 1960; Beeton, 1961 and 1966; Carr, 1962; Thomas,
1963). Beeton (1961) states that there has been a general warning trend in the waters of Lake Erie.
-------
33
Phosphates and nitrates have been charged as the major factors producing accelerated
eutrophication in Lake Erie. Direct measurements, input rates and increased levels have been
recorded (Lewis, 1906; Chandler and Weeks, 1945; Wright, 1933; Curl, 1957; Beeton, 1961;
Kramer, 1961; Matheson, 1962; Verduin, 1964; Harlow, 1966; Ownbey and Kee, 1967).
The only ionic component in Lake Winnipeg showing an increase between 1930 and the period
1962 to 1969 was calcium (Table 2). Values for calcium, iron, chloride and total solids were higher
in 1969 than in 1930 and maximum values for most components were recorded in 1966 and 1967
(Appendix 1). Seichi disc transparencies have declined from one to two meters in 1930 to less than
one meter in 1969 (Bajkov, 1930). Due to the shallow depth of the basin and constant mixing by
wind action, no oxygen depletion results. The variation in chemical concentrations would appear
to be more closely correlated with climatic factors (i.e., precipitation, discharge rates and lake
levels) than loading rates for pollutants (Fig. 4). The absence of any apparent increases in the
concentrations of chemical components in Lake Winnipeg is probably due to the length of
residence time and discharge rates. Residence time for Lake Winnipeg has been estimated at from
four to ten years (Personal communication. Dr. G. Brunskill, Freshwater Institute, Fisheries
Research Board of Canada, Winnipeg). Approximately 30,600 tons of total dissolved solids are
discharged from Lake Winnipeg each day. This value is based on a mean T.D.S. value of 177 ppm
and a mean flow of 59,200 cfs on the Nelson River.
Benthos
Evidence for the eutrophic nature of Lake Winnipeg is provided by both chemical data and
benthos (Tables 2 and 4). Mean standing crop of benthos was calculated as 65.7 ± 3.93 kg/ha.
When this is corrected to deduct the weight of mollusc shells, the value is 58.43 kg/ha. Lakes lying
on the margin of the Precambrian Shield have standing crops of from 7.1 to 9.1 kg/ha (Rawson,
1960). Northern oligotrophic lakes have standing crops ranging from 1.6 to 4.7 kg/ha while lakes
south of the Precambrian Shield normally have standing crops greater than 9.1 kg/ha (Table 4).
The standing crop of benthos in Lake Winnipeg more closely approximates values determined by
Lundbeck (1926) and guarantees its classification as eutrophic. The disparity between values based
on Bajkov's data (1930) and present calculations could be indicative of eutrophication. Areas
sampled and the number of samples probably explain the lower values obtained by Slack (1967).
SO rtor Mean 1918-1967 • 7I3SZ
7O9\.
I93O
1940
I95O
I960
1969
Y«ar
FIGURE 4 Hydrographic data. Lake Winnipeg 1930-1969
-------
34
The densities of benthos in 1930 and the period 1962 to 1969 provide additional proof of
eutrophication. Figures based on Bajkov's (1930) data indicate that benthic densities varied from
437 to 578 organisms/m2. Present data indicates densities of 776 organisms/m2 (Table 3).
In 1930, the amphipods formed from 44% to 85% of total benthic numbers based on certain
stations which correspond to the present sampling program (Bajkov, 1930). This feature Lake
Winnipeg shared with Lake Athabasca and Great Slave Lake (Larkin, 1948; Rawson, 1953). At
these same stations the mayflies formed 40% and the midge larvae 6% of total benthic numbers.
Oligochaetes did not contribute significantly to total benthic numbers. The present data indicates
a decline in amphipod numbers. Decline in both the amphipods and chironomids seems to be
correlated with an increase in both the chironomids and oligochaetes.
The effects of pollutional substances on benthos is manifested by a decline in the numbers of
species, a decline in the densities and numbers of those species intolerant of pollution, and a
corresponding increase in the numbers of those tolerant species (Keup, Ingram and Mackenthun,
1966). Both the amphipods and mayflies are forms known to be intolerant of pollution; their use
as biological indicators of water quality is recognized (Wright and Tidd, 1933; Campbell, 1939;
TABLE 4
Characteristics of Certain Canadian Lakes.
Lake
Gree1
Wollaston1
Reindeer1
Athababasca2
Great Bear3
Great Slave8
Lac la Ronge1 •*
Amisk
Athapapuskow7 '6
Reed7'6
Winnipeg6
Winnipeg7'6
Falcon7'6
Waskesiu1
He a la Crosse1
Area km2
1,152
2,062
5,569
7,770
28,490
27,200
1,178
321
252
190
23,906
Mean Depth (m)
14.9
20.6
17 +
26.
>100.
62
12.7
13.2
30. 9
-
13.9
15
70
446
7.69
11.1
8.2
Total Dissolved
T D S ppm Solids
32
31
61
58
99
150
179
100
109
177
78
188
172
Dry Weight of
Benthos kg/ha
1.6
4.7
1.6
4.1
0.5
2.5
2.8s
9.1
3.4 ± 1.4
2.4 ±1.0
44.1
31.2 ±12.5
3.6 ±1.5
24.6
9.0
1 - Rawson, 1960
5 = Oliver, 1960
9 = approximate
2 - Larkin 4 = Rawson, 1951
6 = Bajkov, 1930 8 = Rawson, 1953
10 = Estimated
-------
35
Gaufin and Tarzwell, 1952 and 1956; Britt, 1955a and b; Gaufin, 1957 and 1959; Beeton, 1961;
Leonard, 1962; Wood, 1963; Beak, 1964; Fremling, 1964; Carr and Hiltunen. 1965; Henson,
1966; Hiltunen, 1969; Grimls, 1969; Arnold, 1969). In the south basin of Lake Winnipeg, the
amphipods have declined from a mean value of 65% to less than 1% of total benthic numbers and
the mayflies from 30% to 13% between 1930 and the period 1962 to 1969 (Bajkov, 1930; Table
3). Even more striking is the absence of mayfly nymphs from grid lines A, B and C (Fig. 2). This
would seem to mark the point of most northerly penetration of pollutants from the Red River.
The Oligochaeta, particularly members of the family Tubificidae, are known to thrive under
conditions of organic enrichment commonly encountered in polluted environments. Oligochaete
densities and species representation have been used as indices of pollution (Wright, 1933; Britt,
1955a and b; Beeton, 1961; Carr and Hiltunen, 1963; Henson, 1966; Johnson and Matheson,
1968; Hiltunen, 1969). Oligochaete densities in the south basin of Lake Winnipeg are normally
larger than the 1400/m2 along the A grid line and decrease to 275/m2 along the B grid line (Fig.
2). Using Wright's system (1933), the density of 1400/m2 would be indicative of moderate
pollution.
Approximately 95% to 98% of all oligochaetes in the south basin of Lake Winnipeg were the
TABLE 4 (Cont'd)
Characteristics of Certain Canadian Lakes.
Lake
Gree1
Wollaston1
Reindeer *
Athabasca2
Great Bear3
Great Slave3
Lac la Ronge1'4
Arnisk1
Athapapuskow7 '6
Reed7'6
Winnipeg6
Winnipeg7 '6
Falcon7'6
Waskesiu1
De a la Crosse1
Major Benthic Groups
Chironomidae
Amphipoda
Amphipoda
Amphipoda and
Chironomidae
Mollusca
Amphipoda
Chironomidae
Amphipoda
Amphipoda
Chironomidae
Amphipoda
Amphipoda and
Epnemeroptera
Chironomidae
Chironomidae
Chironomidae,
Amphipoda and
Sphaenidae
Geology of Basin
Precambrian
Precambrian
Precambrian
Precambrian
Precambrian and
Paleozoic
Precambrian
Precambrian
Precambrian
Precambrian
Precambrian
Precambrian
Precambrian
Precambriam
Paleozoic
Paleozoic
Flushing Time (Years)
14
11
19
3
4 toll 10
0.75
-------
36
species Limnodrilus hoffmeisteri (Claparede),Li'mnodri/ussp. and Tubifex tub if ex (Muller). These
commonly occur in polluted zones (Johnson and Matheson, 1968; Hiltunen, 1969).
The last group of benthic organisms commonly occurring in Lake Winnipeg and known to be
pollution-tolerant is the Chironomidae (Britt, 1955a and b; Beeton, 1960 and 1961; Wood, 1963;
Carr and Hiltunen, 1965; Keup, Ingram and Mackenthun, 1965; Hiltunen, 1969). Brenniman,
Soyak and Curry (in press) have classified chironomic species according to water quality
preference. This method was adopted and modified for use in Lake Winnipeg. All species of the
genera Chironomus, Chironomus (Kiefferulus), Chironomus (Einfeldia) and Glyptotendis were
classified as "pollutional". The genera Pentaneura, Polypedilum, Harnischia, Calopsectra and
Cryptochironomus were classified as cosmopolitan and the remainder of the species as "others",
i.e.. their water quality preference is unknown. No species classified as "clean water" by Curry
were identified in Lake Winnipeg: There is a distinct possibility that species classified as "others"
may, in reality, be clean water forms. However, our data indicates that 39% of the chironomids
could be classified as "pollutional".
The trophic level for Lake Winnipeg was tentatively determined as 1.5 (Brinkhurst, Hamilton and
Herrington, 1968). This is less than that determined for western lake Erie (2.0).
TABLE 5
Classification of Chironomidae According to Water Quality Preference, Lake Winnipeg South Basin,
September 1962 to September 1969 (values are percents of total chironomid numbers for that period)
Year Month Pollutional Cosmopolitan Others
1962 September 54 25 20
1963 March 34 46 20
September 26 53 21
1964 March 17 71 12
September 79 14 7
1965 March 4 66 30
September 43 44 13
1966 March 26 44 30
September 66 20 14
1967 March 43 45 12
September 55 31 14
1968 March 27 67 6
September 22 42 36
1969 March 44 48 8
September 40 50 10
Mean 39 44 17
-------
37
In conclusion, it is evident by changes in the benthic composition, that Lake Winnipeg has been
subjected to eutrophication believed to be the result of pollution. These changes are similar to
those which have occurred in Lake Erie and which are presently occurring in Lake Michigan
(Powers and Robertson, 1965; and Robertson and Alley, 1966). The first change in benthic
composition was a shift from Amphipoda to Chironomidae. Ultimately the benthic population will
be dominated by the Oligochaeta.
Fish Production
It has been noted that the annual commercial fish production in Lake Winnipeg declined in 1961
and has since remained at levels below the 38-year average (Fig. 3). Overfishing and the use of
illegal mesh nets were believed to be the major factors producing the decline (Personal
communication, Mr. W. Pollard, Manitoba Department of Mines and Natural Resources). The
closure of commercial fishing operations on Lake Winnipeg may permit some recovery of fish
stocks.
Hydrographic Data
Since 1961, there has been an increase in Lake Winnipeg water levels terminating in the 1966
maximum (Fig. 4). Increased precipitation and heavy run-off combined to produce high lake levels
accompanied by high chemical concentrations (Table 2). Increased nutrient loading produced high
benthic densities (Table 3). Benthic densities more than four times greater than the value of
801/m2 in September 1966 have been subsequently recorded. While population increases could
have occurred, the concentration method cannot be overlooked as a causative agent.
SUMMARY AND CONCLUSIONS
Between March 1962 and September 1969, surveys on the south basin of Lake Winnipeg were
conducted twice yearly. The purpose of these surveys was to determine changes within the basin
attributable to pollution. Emphasis was directed to the chemical composition of water and
densities and composition of the benthos.
The only ionic component showing an increase since 1930 is calcium. Mean concentrations of
calcium, iron, sulphate and total solids were higher in 1969 than in 1930. Maximum levels for
most chemical components were recorded in 1966 and 1967, corresponding to high lake water
levels. The lack of any significant increases in concentrations is believed to be due to low residence
time and effective flushing rates.
Since 1930, benthic densities have increased from values of 437 and 578/m2 to 776/m2 during the
1962-1969 period. The composition of the benthos has shifted from that dominated by
Amphipoda and Ephemeroptera in 1930 to the present Chironomidae - Oligochaeta configuration.
Approximately 39% of all chironomids and 95% of all oligochaetes, between 1962 and 1969, were
composed of species which could be termed pollutional. As enrichment proceeds, the benthos in
the south basin of Lake Winnipeg will ultimately be dominated by oligochaetes.
ACKNOWLEDGMENTS
Since its inception in 1962, numerous persons have contributed to the pollution monitoring
-------
38
program on Lake Winnipeg. To all, I should like to extend my appreciation and thanks. I am
grateful to Dr. K. H. Doan, Director of Research, Manitoba Department of Mines and Natural
Resources for advocating the preparation of this manuscript and to Dr. F. J. Ward, Professor,
Department of Zoology, University of Manitoba for his constructive criticism.
REFERENCES
Anderson, R. 0. (1959) A modified flotation technique for sorting bottom fauna samples, Limnol.
Oceanogr. 4(2):223-224.
Anon., (1965) Standard methods for the examination of water and wastewater, Twelfth edition,
American Public Health Association Inc., New York, N. Y. 769 p.
Anon., (1969a) High water on Lake Winnipeg, Inland Waters Branch, Dept. Energy, Mines and
Resources, Ottawa, 10 p.
Anon., (1969b) Annual Report, Year ending March 31, 1969, Dept. Mines Nat. Res., Province of
Manitoba, 130 p.
Arnold, D. E. (1969) The ecological decline of Lake Erie, N. Y. Fish and Game J., 16(1):27-45.
Bajkov, A. (1930) Biological conditions of Manitoba lakes, Contributions to Canadian Biology and
Fisheries, Vol. V No. 12:165-205.
Beak, T. W. (1964) A biotic index of polluted streams and its relationship to fisheries, Second
Inter. Conf. Water Poll. Res., Sec. I, Paper No. 10.
Beck, W. M. Jr. and Beck, E. C. (1966) Chironomidae (Diptera) of Florida I. Pentaneurini
(Tanypodinae), Bull. Florida State Museum, Biol. Sci. Vol. 10, No. 8:305-379.
Beeton, A. M. (1960) Great Lakes limnological investigations, Proc. Third Conf. Great Lakes Res.,
Present status and future needs; Mich. Inst. Sci. Tech., Great Lakes Res. Div., Publ. No. 4, p.
123-128.
Beeton, A. M. (1961) Environmental changes in Lake Erie, Trans. Amer. Fish. Soc.,
90(2): 153-159.
Beeton, A. M. (1963) Limnological Survey of Lake Erie, 1959 and 1960, Great Lakes Fishery
Commission Tech. Rept. No. 6, 32 p.
Beeton, A. M. (1966) Indices of Great Lakes eutrophication. Great Lakes Res. Div. Univ. Mich.,
Publ. 15:1-8.
Birkett, L. (1957) Flotation technique for sorting grab samples, J. Cons. Int. Explor. Mer.,
22(3):289-292.
Brenniman, G. R., Soyak, J. A. and Curry, L. L., The distribution of midge larvae (Tendipedidae :
Diptera) as an index of eutrophication in Lake Michigan, In press.
-------
39
Brinkhurst, R. O. (1963) A guide for the identification of British aquatic Oligochaeta, Freshwater
Biol. Assoc. Sci. Pub. No. 22, 52 p.
Brinkhurst, R. O. (1965) Studies on the North American aquatic Oligochaeta II, Tubificidae Proc.
Acad. Nat. Sci., Philadelphia, Vol. 117, No. 4:117-172.
Brinkhurst, R. O. and Cook, D. (1966) Studies on the North American aquatic Oligochaeta III,
Lumbriculidae and additional notes and records of other families. In Proc. Acad. Nat. Sci.,
Philadelphia, 118(1 ):1-33.
Brinkhurst, R. O., Hamilton, A. L. and Herrington, H. B. (1968) Components of the bottom fauna
of the St. Lawrence, Great Lakes, Great Lakes Inst., Univ. Toronto, No. PR 33, 49 p.
Britt, N. W. (1955a) Stratification in western Lake Erie in summer of 1953; effects on the
Hexagenia (Ephemeroptera) population. Ecology, 36:239-244.
Britt, N. W. (1955b) Hexagenia (Ephemeroptera) population recovery in western Lake Erie
following the 1953 catastrophe. Ecology, 36:520-522.
Brown, E. H. (1953) Survey of the bottom fauna at the mouths of ten Lake Erie south shore
rivers; its abundance, composition and use as an index of pollution, Ohio Dept. Nat. Res.,
Division of Water, Lake Erie Pollution Survey Final Report: 156-169.
Butler, G. E. (1949) The lakes and lake fisheries of Manitoba, Trans. Amer. Fish. Soc.,
Seventy-ninth Annual Meeting, Winnipeg: 18-29.
Campbell, M. S. A. (1939) Biological indicators of intensity of stream pollution. Sewage Works J.,
II (1):123-127.
Carr, J. F. (1962) Dissolved oxygen in Lake Erie, past and present. Great Lakes Res. Div., Univ.
Mich. Pub. No. 9:1-14.
Carr, J. F. and Hiltunen, J. K. (1965) Changes in the bottom fauna of western Lake Erie from
1930 to 1961, Limnol. Oceanogr., 10(4):551-569.
Chandler, D. C. (1940) Limnological studies of western Lake Erie, I. Plankton and certain physical
- chemical data of the Bass Island region from September, 1938 to November, 1939, Ohio J.
Sci., 40:291-336.
Chandler, D. C. (1942a) Limnological studies of western Lake Erie, III, Phytoplankton and
physical-chemical data from November, 1939 to November, 1940, Ohio J. Sci., 42:24-44.
Chandler, D. C. (1942b) Limnologicat studies of western Lake Erie, III, Phytoplankton and
physical-chemical data from November 1939, to November, 1940, Ohio J. Sci., 42:24-44.
Chandler, D. C. (1944) Limnological studies of Western Lake Erie, IV, Relationship of
limnological and climatic factors to the phytoplankton of 1941. Trans. Amer. Microscop.
Soc., 43:203-236.
-------
40
Chandler, D. C. and Weeks, O. B. (1945) Limnological studies of western Lake Erie, V. Relation of
limnological and meterological conditions to the production of phytoplankton in 1942, Ecol
Monogr., 15:435-456.
Cober, J. M. E. (1966) Changes in the benthic fauna of the south basin of Lake Winnipeg since
1930 and an assessment of pollution as a possible cause for these changes, M. S. Thesis
unpubl., Univ. Manitoba, 94 p.
Curl, H. C. (1957) A source of phosphorus for the western basin of Lake Erie, Limnol. Oceanogr.,
2(4) :315-320.
Curry, L. L. (1954) Notes on the ecology of the midge fauna (Diptera:Tendipedidae) of Hunt
Creek, Montmorency County, Michigan, Ecology 35(4) :541-550.
Curry, L. L. (1958) Larvae and pupae of the species of Cryptochironomus (Diptera) in Michigan,
Limnol. Oceanogr., 3(4):427-442.
Curry, L. L. (1962) A study of the ecology and taxonomy of freshwater midges
(Diptera:Tendipedidae) of Michigan with special reference to their role in the turnover of
radioactive substances in the hydrosol. Progress Report (Report 2), Central Michigan
University, Dept. Biol., Mount Pleasant, Michigan, 149 p.
Dole, R. B. (1909) The quality of surface waters in the United States, Part I, Analyses of waters
east of the one hundredth meridian, U. S. Geol. Surv., Water Supply Paper 236, 123 p.
Edmondson, W. T., Anderson, C. C. and Peterson, D. R. (1956) Artificial eutrophication of Lake
Washington, Limnol. Oceanogr., 1:47-53.
Einarsson, E. and Lowe, A. B. (1968) Seiches and set-up on Lake Winnipeg, Limnol. Oceanogr.,
13(2): 257-271.
Elson, J. A. (1958) Pleistocene history of southwestern Manitoba, North Dakota Geol. Surv. Misc.
Series, No. 10:62-73.
Fedoruk, A. N. (1964) Studies on the biology, ecology and dynamics of profundal Chironomidae
in Loch Lomond, Ph.D. Thesis, Univ. Glasgow, 198 p.
Fish, C. J. and Associates (1960) Limnological survey of eastern and central Lake Erie, 1928 -
1929, U. S. Fish and Wildl. Sen/., Spec. Sci. Rept. - Fish No. 334, 198 p.
Fremling, C. R. (1961) Screened pail for sifting bottom - fauna samples, Limnol. Oceanogr., 6(1),
P- 96.
Fremling, C. R. (1964) Mayfly distribution indicates water quality on upper Mississippi River,
Science, 146:1164-1166.
Gaufin, A. R. (1957) The use and value of aquatic insects as indicators of organic enrichment.
Biological Problems in Water Pollution, Transaction of a Seminar on Biological Problems in
-------
41
Water Pollution, Robert A. Taft, Sanitary Engineering Center, Cincinnati, Ohio, April 23-27,
1956, U. S. Dept. H.E.W., P.H.S., Bur. Stat. Ser. 136-143.
Gaufin, A. R. (1959) Biological indicators of organic enrichment in fresh water, Biological
Problems in Water Pollution, Transactions of the Second Seminar, April 20-24, 1959,
Cincinnati, Ohio, U. S. Dept. H.E.W., P.H.S., Robert A. Taft, Sanitary Engineering Center
Tech. Rept. W60-3, p. 237.
Gaufin, A. R. and Tarzwell, C. M. (1952) Aquatic invertebrates as indicators of stream pollution,
U.S.P.H.S., Rep. 67(1 ):57-64.
Gaufin, A. R. and Tarzwell, C. M. (1956) Aquatic macro-invertebrate communities as indicators of
pollution in Lytle Creek, Sewage Industr. Wastes, 28:906-924.
Grimas, U. (1969) The bottom fauna of Lake Vattern, Central Sweden, and some effects of
eutrophication, Rept. Inst. Freshwat. Res., Drottingholm, No. 49:49-62.
Harlow, C. G. (1966) Major sources of nutrients for algae growth in western Lake Erie, Great
Lakes Res. Div. Univ. Mich. Pub., 15:389-394.
Hasler, A. D. (1947) Eutrophication of lakes by domestic drainage. Ecology, 28:383-395.
Henson, E. B. (1966) A review of Great Lakes benthos research. Great Lakes Res. Div., Univ.
Mich. Pub. 14:37-54.
Hiltunen, J. K. (1969) The benthic macrofauna of Lake Ontario, Great Lakes Fishery Commission,
Tech. Rept. No. 14, Limnological Survey of Lake Ontario: 39-50.
Johnson, M. C. and Matheson, D. H. (1968) Macroinvertebrate communities of the sediments of
Hamilton Bay and adjacent Lake Ontario, Limnol. Oceanogr., 13(1):99-111.
Johannsen, O. A. (1937) Aquatic Diptera, Part IV, Chironomidae: Sub-family Chironominae,
Memoirs Cornell Univ. Agricultural Experiment Station 210:1-52.
Johannsen, O. A., Townes, H. K., Shaw, F. R. and Fisher, E. G. (1952) Guide to the Insects of
Connecticut, Part IV, The Diptera or True Flies, Fifth Fascile Midges and Gnats, Bulletin No.
80, State Geol. Hist. Surv., 255 p.
Keup, L. E., Ingram, W. M. and Mackenthun, K. M. (1966) The role of bottom-dwelling
macrofauna in water pollution investigations, U. S. Dept. H.E.W., P.H.S., Environmental
Health Series, Water Supply and Pollution Control, 23 p.
Kramer, J. R. (1961) Chemistry of Lake Erie, Great Lakes Res. Div., Univ. Mich. Publ. No.
7:27-56.
Ladell, W. R. S. (1936) A new apparatus for separating insects and other anthropods from the soil,
Ann. Appl. Biol.. 23:862-879.
-------
42
Larkin, P. A. (1948) Pontoporeia and Mysis in Athabaska, Great Bear and Great Slave Lakes, Bull.
Fish. Res. Bd. Canada, 78:33 p.
Larkin, P. A. (1964) Canadian lakes. Verb. Internat. Verein, Limnol.. XV:76-90.
Leonard, J. W. (1962) Environmental requirements of Ephemeroptera, Biological Problems in
Water Pollution, Third Seminar, August 13-17, 1962, Cincinnati, Ohio, Environmental Health
Series, Water Supply and Pollution Control, U. S. Dept. H.E.W., P.H.S.: 110-117.
Leonard, J. W. and Leonard, F. A. (1962) Mayflies of Michigan trout streams, Cranbrook Institute
of Science, Bloomfield Hills, Michigan, 139 p.
Leverin, H. A. (1947) Industrial waters of Canada, Report on investigations, 1934 to 1943,
Canadian Dept. Mines Res., Mines Geol. Br., Rept. No. 819, 109 p.
Lewis, S. J. (1906) Quality of water in the upper Ohio River basin and at Erie, Pa., U. S. Geol.
Surv.. Water supply Irrigation Pap. 161, 114 p.
Lundbeck, J. (1926) Die Bodentierwelt norddeutscher Seen, Archiv. Hydrobiol., Supp., 7:1-473.
Lyman, F. E. (1943) A pre-impoundment bottom-fauna study of Watts Bar Reservoir area
(Tennessee), Trans. Amer. Fish. Soc., 72:52-62.
Mason, W. T. Jr. and Yevich, P. P. (1967) The use of Phloxine B and Rose Bengal stains to
facilitate sorting benthic samples, Trans. Amer. Microsc. Soc., 86(2) :221-223.
Mason, W. T. Jr. (1968) An introduction to the identification of chironomid larvae. Division of
Pollution Surveillance, F.W.P.C.A., U. S. Dept. Int., 89 p.
Matheson, D. H. (1962) A pollution study of western Lake Erie, Inst. Sci. Tech., Univ. Mich. Pub.
No. 9:15.
Matheson, D. H. (1965) A pollution study of western Lake Ontario, Proc. Great Lakes Res. Div.,
Univ. Mich. Pub. No. 9:15-20.
Minder, L. (1926) Biologische-chemische Untersuchungen im Zurichsee, Rev. d'Hydrologie,
3:1-69.
Miller, R. B. (1947) Great Bear Lake, Bull. Fish. Res. Bd. Canada, 72:31-44.
Neave, F. (1932) A Study of the May Flies (Hexagenia) of Lake Winnipeg, Contributions to
Canadian Biology and Fisheries, Vol. VII, No. 15 (Series A, General, No. 12):179-201.
Northcote, T. C. and Larkin, P. A. (1963) Western Canada, Chapter 16, p. 451-485, In Limnology
in North America, D. C. Frey, Editor, Univ. Wis. Press, Madison, Wis.
Oliver, D. R. (1960) The macroscopic bottom fauna of Lac la Ronge. Saskatchewan, J. Fish. Res.
Bd. Canada, 17(5) :607-624.
-------
43
Ownbey, C. R. and Kee, D. A. (1967) Chlorides in Lake Erie, Proc. Great Lakes Res. Conf.
10:382-389.
Powers, C. F., Jones, D. L., Mundinger, P. C. and Ayers, J. C. (1959) Exploration of collateral data
potentially applicable to Great Lakes hydrography and fisheries, Final Rept., Phase II, U. S.
Fish and Wildlife Serv., Contract No. 14-19-008-9381, Great Lakes Res. Inst., Univ. Mich.,
164 p.
Powers, C. F. and Robertson, A. (1965) Some quantitative aspects of the macro-benthos of Lake
Michigan, Great Lakes Res. Div., Univ. Mich. Publ., 13:153-159.
Powers, C. F. and Robertson, A. (1966) The aging Great Lakes, Sci. Amer., Nov. 1966:94-100.
Rawson, D. S. (1951) The total mineral content of lake waters. Ecology, 32(4):669-672.
Rawson, D. S. (1953) The bottom fauna of Great Slave Lake, J. Fish. Res. Bd. Canada, Volume X,
No. 8:486-520.
Rawson, D. S. (1960) A limnological comparison of twelve large lakes in northern Saskatchewan,
Limnol. Oceanogr., 5(2): 195-211.
Roback, S. S. (1957) The immature tendipedids of the Philadelphia area, Monog. Acad. Nat. Sci.
Philadelphia. No. 9, 152 p.
Robertson, A. and Alley, W. P. (1966) A comparative study of Lake Michigan macrobenthos,
Limnol. Oceanogr., 11 (4):576-583.
Slack, H. D. (1965) The profundal fauna of Loch Lomond, Scotland, Proc. Ray. Soc., Edinburgh
B, 69:272-297.
Slack, H. D. (1967) A brief survey of the profundai benthic fauna of lakes in Manitoba, J. Fish.
Res. Bd. Canada 24(5): 1017-1033.
Thomas, J. F. J. (1954) Industrial water resources of Canada, Upper St. Lawrence River -Central
Great Lakes drainage basin in Canada, Dept. Mines Tech. Serv., Water Surv. Rept. No. 3,
Mines Br. Rept. No. 837, 212 p.
Thomas, N. A. (1963) Oxygen deficit rates for the central basin of Lake Erie, Great Lakes Res.
Div. Univ. Mich., Pub. 10:133.
Townes, H. K. Jr. (1945) The nearctic species of tendipedini (diptera, tendipedidae-chironomidae),
Amer. Mid. Nat., Vol. 34, No. 1,266 p.
Wand, P. (1966) Lake Winnipeg water drops-damage rises, Winnipeg Free Press, September 6,
1966.
Verduin, J. (1964) Changes in western Lake Erie during the period 1948 - 1962, Verh. Int.
Verein., Limnol., XV:639-644.
-------
44
Ward, B. W. and Whipple, G. C. (1959) Freshwater Biology, 2nd Edition, John Wiley and Sons,
New York, 1248 p.
Wood, K. C. (1963) The bottom fauna of western Lake Erie, Great Lakes Res. Div., Univ. Mich.,
10:259-265.
Wright, S. (1933) Limnological survey of western Lake Erie, Special Scientific Report Fisheries
No. 139, U. S. Dept. Fish and Wildlife, 341 p.
Wright, S. and Tidd, W. (1933) Summary of limnological investigations in western Lake Erie in
1929 and 1930, Trans. Amer. Fish. Soc., 63:271-285.
-------
APPENDIX 1
Ranges in Values for Chemical Data (mg/l). Lake Winnipeg South Basin, March 1962
Physical and
Chemical Factors
True color P+ Co units
1962 1963
March September March September
1964
March September
1965
March September
Apparent color P+ Co units 20 to 50
(35.6)
Turbidity
pH
Calcium
Magnesium
Sodium
Iron
Total hardness
Phosphate (ortho)
Alkalinity
Bicarbonate
Carbonate
Chloride
Carbon dioxide
Dissolved oxygen
B. O. D.
Sulphate
Spec. cond. jU mhos
Totat solids
Suspended solids
Dissolved solids
Number of samples
1.0 to 30
(10.8)
7.0 to 7,8
(7.3)
16.8 to 94.4
(26.3)
-
0.6 to 3. 9
(0.4)
54 to 382
(96.7)
-
18 to 158
(34.4)
•
•
2.0 to 20.4
(6.8)
4.6 to 732
(9.5)
6.5 to 16.9
(13.7)
*w
•
-
38 to 650
(161)
•
40 to 70
(52.5)
4.5 to 34
(15.1)
7.2 to 8.4 6.7 to 7.6
(7.6) (7.3)
16.0 to 40.0
(23.9)
-
-
0.1 to 1.08
(0.5)
54 to 160
(91.3)
-
92 to 226 40 to 194
(144) (99.8)
-
-
2.0 to 13.5
(5.6)
4,0 to 44 1.7 to 6.8
(13.7) (3.2)
8.2 to 11.29.6 to 17.1
(9.4) (13.1)
2. 9 to 8.4
(4.9)
-
-
94 to 284
(56)
1.5 to 27
(10.4)
-
22
-
-
6.6 to 7. 6
(7.2)
14.4 to 36.8
(28.9)
3.9 to 13.6
(10.2)
2.5 to 13.5
(8.2)
0.3 to 2.3
(0.8)
52 to 148
(115)
-
44 to 122
(89.5)
53 to 132
(106)
0 to 4.8
(1.2)
1.0 to 11.0
(6.5)
x
9.2 to 9.8
(9.5)
0.5 to 1.4
(0.9)
8.2 to 40.4
(28.3)
106 to 238
(190)
-
82.0 to 309.9
(189)
24
30 to 60
(48.5)
4.0 to 32.5
(17.1)
6.7 to 8.0
(7.2)
15.2 to 36.0
(25.2)
0.12 to 0.68
(0.34)
54 to 126
(99.6)
•
40 to 181
(104.4)
1.5 to 100
(5.6)
0 to 20.4
(3.9)
3.2 to 17.7
(12.1)
0.4 to 6.3
(3.1)
-
80 to 210
(148)
1.0 to 9.0
(3.9)
32
-
-
7.2 to 8.1
(7.7)
14.4 to 60.0
(25.0)
4.9 to 12.2
(9.0)
2.0 to 9.5
(6.4)
1.4 to 3.9
(2.3)
56 to 122
(94.2)
48 to 94
(75.3)
58 tollS
(91.3)
1.2 to 10.0
(5.9)
-
-
11.6 to 37.8
(28.3)
-
110 to 286
(214)
•
18
"
6 to 20
(10.3)
7.4 to 8.2
(7-8)
0.8 to 36.0
(28)
-
-
66 to 140
(105.8)
-
52 to 116
(87.7)
-
'
2.0 to 13.5
(6.7)
-
8.0 to 17.5
(13.9)
2.0 to 4.1
(3.3)
-
-
102 to 368
(164)
0 to 37
(14.4)
39
-
7.5 to 7.9
(7.6)
22.0 to 32.8
(26.4)
1.7 to 12.6
(9.3)
4.5 to 11.5
(7.6)
0.36 to 1.4
(0.73)
41 to 134
(104)
0.11 to 0.40
(0.21)
40 to 98
(77)
40 to 110
(94.3)
'
4.0 to 11.5
(8.5)
-
28.1 to 49.5
(38.8)
-
150 to 216
(173)
8
-p.
01
-------
APPENDIX 1 (cont'd)
Ranges in Values for Chemical Data (mg/l). Lake Winnipeg South Basin, March 1962
Physical and
Chemical Factors
True color P+ Co units
1966
March September
1967
March September
15 to 25
(19)
1968
March September
15 to 45
(25)
191
March
30 to 40
(36)
59
September
30 to 100
(42)
Apparent color P+ Co units ..-•--•
Turbidity
PH
Calcium
Magnesium
Sodium
Iron
Total hardness
Phosphate (ortho)
Alkalinity
Bicarbonate
Carbonate
Chloride
Carbon dioxide
Dissolved oxygen
B. O. D.
Sulphate
Spec, cond, M mnos
Total solids
Suspended solids
Dissolved solids
Number of samples
-
7.3 to 7.7
(7.5)
14.4 to 36.8
(22.4)
2.9 to 12.6
(7.0)
2.0 to 13.0
(5.2)
0.14 to 0.32
(0.2)
48 to 144
(85.2)
0.11 to 0.22
(0.14)
WP
48.8^129
-
0 to 12
(4.3)
10.2 to 16.5
(14.4)
-
17 to 48
(28)
-
82 to 212
(132)
-
74 to 208
(127)
20
9 to 50
(26)
7.9 to 8.0
(8.0)
17.6 to 44.2
(33.7)
10.9 to 18.7
(14.2)
9.5 to 17.8
(13.3)
0.11 to 1.8
(0.98)
118 to 188
(151)
0.34 to 1.3
(0.53)
90to3139
110 to 170
(138)
0.2 to 13.4
(5.4)
-
7.7 to 10.0
(9.0)
-
16.5 to 58.4
(365)
-
160 to 280
(211)
96 to 276
(183)
8
50
T.8
59.2
32'.1
26.4
14.2
280
0.39
182
222
19
10.0 to 16.
(13.5)
-
45.3
-
642
338
1
4 to 130
(38)
7.4 to 8.2
(7.8)
27.2 to 48.0
(36.2)
9.7 to 21.9
(13.8)
8.6 to 18.8
(11.4)
0.04 to 1.63
(0.44)
112 to 240
(147)
-
-
•
-
11.5 to 19.8
(15.7)
-
,3 7.3 to 10.6
(8.7)
28.2 to 60.0
(41.8)
-
170 to 346
(206)
4 to 64
(14)
158 to 304
(203)
34
15 to 55
(31)
7.98 to 8.11
(8.07)
30.4 to 38.4
(33.2)
7.3 to 12.6
(10.8)
8.3 to 13.0
(10.4)
°-W'7
110 to 144
(126)
0.05 to 0.11
(0.08)
82 to 120
(98)
100 to 146
(119)
-
9.1 to 13.3
(11.0)
4 to 10
(7.3)
0.0 to 10.1
(7.1)
-
28.8 to 41. 2
(35.1)
195 to 290
(229)
-
•
20
5 to 60
(18)
7.33 to 8.20
(7.83)
17.6 to 49.6
(30.0)
0.28 to 19.4
(9.3)
3.2 to 17.6
(9-7)
0.14 to 3.12
(0.59)
57 to 196
(113)
0.0 to 0.36
(0.13)
44 to 144
(86)
53 to 176
(105)
2.5 to 18.0
(8.8)
•
9.8 to 10.4
(9.9)
-
3 to 60
(34)
-
9.0 to 326
(180)
0 to 50
(15)
80 to 290
(160)
32
3 to 30
(16)
"f^7)-72
27.2 to 31. 2
(28.0)
8.9 to 11.2
(9.9)
9.3 to 12.0
(10.1)
0.16 to 1.04
(0.28)
105 to 122
(110)
0.13 to 0.27
(0.17)
76 to 88
(83)
93 to 107
(100)
7,0 to 8.0
(7.3)
1.9 to 32.4
(7.2)
4.5 to 16.0
(12.2)
-
24.7 to 31.3
(29.4)
235 to 270
(252)
150 to 202
(175)
0 to 52
(10)
146 to 180
(165)
10 to 320
(35)
7. 65 to 8.99
(7.98)
12.8 to 52.0
(29.0)
2.9 to 22.8
(11.1)
2.7 to 35.3
(12.1)
0.20 to 7.0
(1.31)
48 to 224
(116)
0.085 toO.44
44t02168
54 to 205
(114)
"
1.5 to 22.8
(8.8)
0.1 to 1.5
(0.3)
8.3 to 9.6
(8.7)
0.4 to 2.3
(1.3)
3.6 to 8.2
(4.0)
110 to 580
(298)
98 to 566
(246)
6 to 62
(22)
92 to 374
(212)
33
-------
47
APPENDIX 2
Larval Forms of Chironomidae Collected
Subfamily Tenypodinae
Tanypus sp.
Procladius sp.
Procladius riparius (Malloch)
Procladius culiciformis (Linnaeus)
Procladius nr. adumbratus (Johannsen)
Coelotanypus sp. (Kieffer)
*Clinotanypus sp. (Kieffer)
•Zavrelimyio sp.
Peritoneum cornea (Fabricius)
Ablabesmyia sp.
Subfamily Diamesinae
*Diamesa sp.
Subfamily Orthocladiinae
•Canliocladius sp. (Kieffer)
'Heterotrissocladius sp.
*Nanocladius sp.
*Orthocladius sp.
*Psectrocladius sp. (Kieffer)
•Trichocladius sp. (Kieffer)
Subfamily Chironominae
Chironomus sp.
Chironomus attenuates (Walker)
Chironomus riparius (Meigen)
Chironomus plumosus (Linnaeus)
Chironomus tentans (Fabricius)
Chironomus tuxis (Curran)
Chironomus staegeri (Lundbeck)
Chironomus anthracinus (Zetterstedt)
*Chironomus neomodestus (Malloch)
*Chironomus tendipediformu (Goetghebuer)
'Chironomuspaganus (Meigen)
^Chironomus hyperboreus (Staeger)
*Chironomus ochreatus (Townes)
Chironomus (Kiefferulus)
Chironomus (Einfeldia)
Microtendipes sp.
Glyptotendipes sp. (Kieffer)
Cryptochironomus digiiatus (Malloch)
Cryptockironomus fulvus (Johannson)
Polypedilum (Tripodura) scalaenum (Schrank)
Chironomus (Cryptochironomus) nais
Cryptochironomus (Hamischia)
Ttmytarsus (Vander Wulp) = Calopsectra (Kieffer)
Hamischia (Kieffer)
Reference
Mason, 1968
Roback, 1957
Mason, 1968
Mason, 1968
Mason, 1968
Curry, 1962
Mason, 1968
Curry, 1962
Roback, 1957
Mason, 1968
Roback, 1957
* The identity of these genera and species is in doubt. Few specimens were obtained and all were
larvae.
-------
EUTROPHICATION IN SOME LAKES AND COASTAL AREAS
IN FINLAND, WITH SPECIAL REFERENCE TO POLYHUMIC LAKES
Pasi O. Lehmusluoto
INTRODUCTION
The dominant feature of the Finnish inshore waters is the fact that the lakes are linked together by
short stretches of running water. They form in this way watercourses (Fig. 1). The number of lakes
is estimated to be within the range of 55,000 • 75,000. Their total area is 32,000 km2 and water
volume is about 220 km3 (mean depth about 7 meters). The mean run-off (MQ) is 95 km3/year.
The average color of the water is comparatively high in the whole country, i.e. 91 mg Pt/l. This
means that the average hurnus content of the water is about 13 mg humus (dry weight) per liter
(Ryhanen, 1968). The iron content is also comparatively high, and averages 1.1 mg/l. The lake
water is normally clear, slightly acid (pH 6.6), and the conductivity is low, under 70 micromhos
(Laaksonen, 1970).
25 E
27'E
29 E
61 N
61 N
25-E
27 E
29 E
FIGURE 1 A part of the Finnish lake district. S=Lake Saimaa.
This investigation was supported in part by the National Research Council for sciences in
Finland, in part by the Foundation for Research of Natural Resources in Finland and in part by
the City of Helsinski.
48
-------
49
Allotrophic humus, and organic matter produced in autotrophic phytoplankton primary
production, especially in connection with eutrophication, have been found to be most important
factors in protection of polyhumic lakes in maintaining sufficient oxygen concentration in the
water bodies (Ryhanen, 1968). In certain areas carbon rich pulp mill effluents may also play an
important role in loading the waters.
The lakes in Finland are normally dimictic. The spring circulation is not, however, always
complete in small lakes. The ice period lasts on the lakes for 150 - 250 days, making the conditions
in the water more favorable for decomposition than for phytoplankton primary production. In the
hypolimnion the oxygen consumption caused by different kinds of oxidation processes (e.g. the
decomposition of organic material produced in primary production, the decomposition of pulp
mill effluents, and ferrous iron oxidation) may be great, and lead to an oxygen deficit during
winter and summer stagnations. This is especially severe during summer stagnations, when the
preceding spring circulation has not been complete.
Increase in phytoplankton primary production is perhaps not as important in polyhumic lakes as
in the nonhumic lakes. This is due to the "inhibitive" effect of humus to the total daily
phytoplankton primary production (Shapiro, 1957). This effect is caused mainly by the strong
light absorption of humic material. In algal assay bottle tests in constant light, humus has shown to
be slightly biostimulative (Demmerle, 1967). The visible light (E 400 - 700 nm) penetrates
polyhumic lakes only 2-3 meters (Fig. 2). Humus alone depresses the trophogenic layer beyond the
normal properties of the water. Thus in polyhumic water bodies there is produced daily only a
part of that phytoplankton material produced in similar nonhumic waters.
MG C/M3/ DAY - %
I 10 IOO IOOO IOOOO
EUTROPHK
10 IOO I IO 100 I
I
2
3-
i-
to-
20
OLK30TROPHC
POLYHUMIC
OLIGOTROPHIC
NON-HUMIC
10 IOO
OLIGOTROPHIC
'BLUE LAKE"
FIGURE 2
Phytoplankton primary production (mg C/m3/day, solid line) and visible light
extinction (%, dashed line) vs. depth in different kinds of waters in the middle of the
growing season. Data for the oiigotrophic "blue lake" is from Rodhe et al. (1966).
Phytoplankton production and light extinction are in logarithmic scale.
-------
50
In some clear water lakes, so called "blue waters", where the optical purity of the water is great
(Fig. 2), the effect of ultraviolet radiation can cause a reverse vertical distribution of
phytoplankton primary production (Rodhe et al., 1966). Ultraviolet radiation depresses this
production below its potential capacity because phytoplankton production can proceed only in
the layers where the influence of ultraviolet radiation is negligible and there is still visible light
illumination. In these lakes this layer is often beyond the optimum visible light layer in the lower
epilimnion (Rodhe et al., 1966).
Normally in nonhumic and nonturbid lakes increasing phytoplankton mass causes the trophogenic
layer to be depressed, and the maximum phytoplankton primary production per cubic meter (in
optimum visible light layer) is moved closer to the water surface. In polyhumic and very turbid
lakes this does not happen because of the already thin trophogenic layer (Fig. 2).
In Finnish lakes, as in some mountain lakes of Austria (Pechlaner, 1964), high phytoplankton
primary production beneath the ice during spring time normally cannot be found. The
phytoplankton primary production is at that time barely measurable. The lack of an under-ice
spring bloom may be partly due to the relatively fast ice melting, as the ice does not remain on the
water long enough after the snow has melted to allow sufficient light for intensive phytoplankton
production. Low rates of primary production can be measured under ice normally during early
winter before the ice is covered with snow (Pechlaner, 1964).
In Alaska almost half of the annual phytoplankton primary production may occur beneath the ice
sheet in spring time, e.g. Lake Schrader and Lake Peters (Hobbie, 1964). The highest
phytoplankton production for the whole year in Lake Schrader was also found in spring under the
ice. Rodhe et al. (1966) measured considerable under-ice phytoplankton production in Swedish
Lapland. The high under-ice production in the two Alaskan lakes in spring time was a result of the
relative lack of turbulence in the water beneath the ice. The algae remained in the euphotic zone
continuously during ice period. After the ice had left and the lakes began to circulate, the
phytoplankton primary production dropped sharply. In Finnish lakes the phytoplankton primary
production usually increases sharply during this period. At this time of year the algae can stay in
the euphotic zone only a small part of their lives, as in polyhumic lakes during summer time. These
algae also appear to be adapted to low light intensities (Hobbie, 1964).
The growing season is short in Finland with daylight in June averaging 18 hours, August 16 hours,
but in October only 10 hours. The ice period normally lasts from November to April.
The vernal diatom bloom is dependent on the length of the day and on the illumination during the
day (Gran, 1929). This bloom can be normally found in the Finnish waters from the end of April
to the end of May, when the solar radiation exceeds 2.5 - 3.5 kcal/m2/min. The vernal diatom
bloom could possibly be found when the solar radiation exceeds 1.5 kcal/m2/min (Smayda, 1959),
but the bloom is hindered at that time by the ice cover (Lehmusluoto, 1968).
During summer time in polyhumic lakes, the lack of light-and in some Alaskan and Swedish
mountain lakes, the overwhelming light (Hobbie, 1964, Rodhe et al., 1966) seems to limit the
total daily primary production per square meter of the water, but the nutrients may limit the
maximum primary production of a cubic meter of water (Hobbie, 1964).
When discussing primary production, one cannot help dealing with the lake typology. Nowadays it
-------
51
has been suggested that primary production could be the basis for the trophic classification of
waters (Rodhe, 1958, 1969, Findenegg, 1964). There are many parameters of primary production
that could be used: 1) the shape of the primary production curve {Findenegg, 1964), 2) primary
production per square unit (diurnal, annual) (Rodhe, 1958, Findenegg, 1964, Hubel, 1968), 3)
maximum primary production per unit volume (Rodhe, 1958, Lehmusluoto, 1969) 4) so called
V/0-quotient as suggested by Rodhe (1958).
I would like to suggest that the mean maximum primary production rate per unit volume in the
water column during the growing season measured in situ, or in constant light, could be used as an
index for many kinds of water bodies. This would give a relative but objective index for the
phototrophic level of the water. It would not reflect the total amount of organic material built up
in the whole water column (which can be estimated by the light extinction curve), but it might be
informative about the degree of eutrophication of the water. No suggestions for the designation of
oligotrophic or eutrophic waters are presented, because the data available is insufficient.
When talking about lake typology, especially of polyhumic waters, it may be appropriate to use
phytoplankton primary production only as a measure of the autotrophic state. The direct
comparison of trophic states of humic and nonhumic lakes is impossible, as the classification of
polyhumic lakes on the basis of phytoplankton production cannot give an objective result
(Ryhanen, 1968). Ohie (1940, 1956) has suggested that also the heterotrophic state should be
taken into account.
EXAMPLES OF EUTROPHICATION
Experiments with Polyhumic Water
In order to get some information about the role of nitrate nitrogen and phosphate phosphorus in
the eutrophication process of polyhumic waters, a series of preliminary tests were made by algal
assay bottle tests (Bringmann and Kuhn, 1956; Skulberg, 1964), with water from the polyhumic
Lake Hakojarvi (61° 15' N, 25° 12' E). The color of the water was about 150 mg Pt/l and the iron
content averaged in the whole water column (maximum depth 15 meters) about 0.5 mg/l. The ice
period lasts on this lake about 180 days, from November to the end of April. Spring circulation
was complete during 1966 - 1970 only twice.
Results showed that nitrogen was the primary limiting nutrient. Nitrogen was also a limiting
nutrient for bacterial growth in this water (Sederholm, 1969). When nitrogen and phosphorus were
added together a large algal growth occurred (Fig. 3).
A series of experiments on phytoplankton primary production during 1968-1969 using 7 in situ
plastic test cells were conducted. The cells were 10 meters deep, 1.2 meters in diameter, and the
top was open (Goldman, 1962). The water volume was about 12 m3. Only the results of the test
cells of 1968 are dealt with in this paper, as the results of 1969 were quite similar in nature.
Nitrate nitrogen and phosphate phosphorus were added to the test cells according to the
preliminary tests by algal assay in bottles. Nutrient additions varied from 0.05 to 1.0 mg/l nitrogen
and 0.005 - 0.1 mg/l phosphorus. The nutrient additions to the different test cells are shown in
Table 1.
-------
52
AU3AL ASSAY
CELLS/LITER
rSIO10
- 4-I010
10'°
I MS m_ (ADDED)
FIGURE 3 The influence of different nitrogen (nitrate) and phosphorus (phosphate) additions in
polyhumic lake water (Lake Hakojarvi) as measured by algal assay in bottles. Test
alga was Chlorella sp.
CeD No.
NO3-N,mg/l
PO4-P,mg/l
TABLE 1
Nutrient Additions into the Test Cells
1234 5 67
1.0 - 1.0 0.5 0.1 0.05
0.1 0.1 0.05 0.01 0.005
Phytoplankton primary production, which was measured by the carbon-14 method (Steemann
Nielsen, 1952), was made eleven times during the test period, (17.7. - 8.10. 1968). The test period
was about 2/3 of the whole growing season. The background production (Cell No. 1) was 1.58 g
C/m2/test period (average for test period 20.2 mg C/m3 (max.)/day). In Lake Hakojarvi in
1966-1969 the phytoplankton primary production averaged 2.66 g C/m2 /growing season (average
for test period 17.7 mg C/m3 (max.)/day). There were not any large changes in phytoplankton
production in the test cell, where only nitrogen had been added, but phosphorus addition alone
caused a slight increase (Table 2). Phosphorus did not alone cause significant eutrophication, i.e.
phosphatetrophication (Thomas, 1968). In the cells where both nitrogen and phosphorus were
added the increase in phytoplankton production was greater with greater .additions of nutrients.
The highest value. 8.16 g C/m2/test period (154.2 mg C/m3 (max)/day in test period), was in the
cell containing the greatest nutrient addition (1.0 mg N/l and 0.1 mg P/l).
-------
53
1
1.58
29.2
2
1.32
23.5
3
1.42
44.1
4
8.16
154.2
5
3.48
65.3
6
2.80
49.6
7
1.92
51.5
TABLE 2
Phytoplankton Primary Production g C/m2/test period and mg C/m3 (max.)/day
on the Average in the Test Period For Nutrient Additions See Table 1
Cell No.
g C/m2 /test period
mgC/m3 (max.)/day*
* Average for the growing season (eleven measurements).
The oxygen stratification in the different test cells did not show any great variations. In the
epilimnion at a depth of 1 meter each cell contained a range of 8.3 to 8.9 mg Oj/l, and at 5 meters
depth from 6.8 to 7.1 mg O2/l. in the hypolimnion at 10 meters depth between 5.7 to 6.2 mg
02/l. The highest oxygen values in the epilimnion were normally found in the cells where both
nitrogen and phosphorus had been added. However, the lowest values in hypolimnion were not
found in these cells. Nutrient additions in these experiments apparently did not stimulate oxygen
consumption in the hypolimnion.
The uptake of radioactive glucose by heterotrophic bacteria (Wright and Hobbie, 1965) was also
stimulated by nitrogen and phosphorus. In the test cell where the greatest nitrogen and
phosphorus addition was made, primary production increased about five times, and the uptake of
radioactive glucose increased about ten times (Leppanen, 1970). The bulk of heterotrophic
bacteria which caused the uptake of radioactive glucose may have been dependent on the algal
mass produced, because the increase of glucose uptake occurred only in the upper epilimnion
where primary production was at a maximum (Fonden, 1969). The same nutrient addition was
present both in the lower epilimnion and in hypolimnion.
Although it seems that the humus in the water did not in these experiments form so serious a
problem in maintaining sufficient oxygen concentration in hypolimnion as the organic material
built up by the phytoplankton primary production, humus may be important in the total
production of polyhumic waters. The zooplankton and fish production in the Finnish polyhumic
lakes seem to be greater than the phytoplankton primary production can yield (Ryhanen, 1968).
It is proposed that humus as such, or transformed to bacterial biomass, may serve as food for the
zooplankton (Jarnefelt, 1956), and thus lead to a relatively high fish production. It has been
shown that the bacterial numbers in humic waters are normally comparable to the numbers in
eutrophic lakes (McCoy and Sarles, 1969).
On March 22, 1970 the entire Lake Hakojarvi was fertilized by adding 750,000 m3 phosphorus
and nitrogen to bring the lake water concentration up by 1 mg N/l and 0.1 mg P/l. Results from
this experiment are not yet available.
The above experiments show the role of nitrogen and phosphorus in the eutrophication process in
the water of Lake Hakojarvi. In order to get some information about eutrophicative effects of
domestic sewage and pulp mill effluents, which are typical pollutants in Finland, some data is
presented in the following from Lake Saimaa (Fig. 1) and two coastal areas in the Gulf of Finland.
-------
54
Lake Water
In Lake Saimaa, which is one of the largest lakes in Finland, phytoplankton production was
studied in two areas during 1968. The first area in northern parts of Lake Saimaa was being
polluted by about 12,000 m3/day of biologically purified domestic sewage. The second area was in
the south, and was polluted by 260,000 m3/day of mechanically purified pulp mill waste. The
effluent consists of 44% sulphite and 56% sulphate liquor (Lehmusluoto and Heinonen, 1970).
Preliminary tests were made by using algal assay bottle tests mentioned earlier.
Domestic sewage biostimulated the algae growth in every concentration, but pulp mill effluents
were almost lethal to algae at 10% effluent concentrations. In 1% and 0.1% effluent
concentrations, algae growth was slightly stimulated {Fig. 4).
Primary production measurements from the lake water were made six times during the growing
season in constant light (5,000 lux) to eliminate the diurnal changes in illumination.
Domestic sewage caused increased primary production in the recipient. Phytoplankton production,
without any depression near the sewage outfall, decreased, as did the oxygen consumption in
ALGAL ASSAY
CELLS/LITER
- . DOMESTIC SEWAGE
PULP MILL EFFLUENTS
(SA + SI)
I06-
100 50 25 10 I 0.1 0
WASTE WATER CONCENTRATION IN LAKE WATER %
FIGURE 4 The influence of domestic sewage and pulp mill effluents (sulphate, sa and sulphite,
si liquor) in lake water (Lake Saimaa) as measured by algal assay in bottles. Test alga
was Ankistrodesmus sp.
hypolimnion with distance. In the hypolimnion there was no oxygen deficit in any region. The
greatest values of phytoplankton production (294.1 and 299.5 mg C/m3/day) were observed near
the outfall. At a distance of about 24 km from the outfall, the primary production was 48.1 mg
C/m3/day (Table 3). This value can be used as an index for unpolluted water in that region.
-------
55
TABLE 3
Phytoplankton Primary Production mg C/m3/day in Northern Parts of Lake Saimaa
as Mean Values of Six Measurements During the Growing Season
The Recipient is Influenced by Domestic Sewage
Distance from
Outfall, km 248 12 24
mgC/m3/day 294.1 299.5 239.7 170.4 48.1
The pulp mill effluents seemed at first to hinder phytoplankton primary production, but caused a
pronounced eutrophication (752.9 mg C/m3/day) about 9 km from the outfall. At a distance of
about 22 km, phytoplankton production was 175.2 mg C/m3/day (Table 4), and the water was
still slightly eutrophic.
It is shown that both domestic sewage and pulp mill effluents did cause eutrophication of the
recipient. Sewage did not have an inhibitive effect as did the effluents. The maximum
eutrophication caused by the pulp mill effluents was more intensive than that caused by the
domestic sewage, but it occurred far from the discharge area.
TABLE 4
Phytoplankton Primary Production mg C/m3/day in Southern Parts of Lake Saimaa
as Mean Values of Six Measurements During the Growing Season
The Recipient is Influenced by Domestic Sewage
Distance from
Outfall, km -4 2 4 9 13 22
mgC/m3/day 93.1 177.9 285.6 752.9 548.8 175.2
Coastal Water
Eutrophication is not only a problem of lakes but also of some coastal areas. Coastal waters in
Finland are brackish waters, partly meso- and oligohatine. Problems are concentrated usually near
the cities on the coast.
Helsinki, the capital of Finland, initiated in 1965 a research program on the pollution of the Baltic
around the city. This was initiated because the archipelago near the city is an important recreation
area for the citizens during the summer time and the quality of the water was decreasing with
increasing eutrophication. This was mainly due to blue-green algae blooms in the later summer.
The City of Helsinki was discharging its mechanically and biologically purified waste waters into
the nearby bays. These purification procedures seemed inadequate to control the eutrophication in
this coastal area.
-------
56
One part of the research program consisted of phytoplankton primary production studies. In this
paper the results of one of the fjord systems of the year 1967 will be briefly dealt with.
Phytoplankton production, mg C/m3 (max.)/day, in the middle of the growing season at 1 km
from the outfall was about 30 times higher than that at 12 km distance in the unpolluted area
outside Helsinki (Fig. 5). On the average during the growing season it was about 20 times higher
(Table 5). Fluctuations in primary production per square meter became irregular with increasing
eutrophication (Fig. 6). Annual primary production in the innermost bay was 174 g C/m2/year,
i.e. over 5 times higher than in the unpolluted areas in the Gulf of Finland where it was 30 g
C/m2/year (Table 5).
TABLE 5
Primary Production g C/m2 /Growing Season and mg C/m3 (max.)/day on the Average
in the Growing Season in a Fjord System in the Gulf of Finland off the Coast of Helsinki in 1967
Distance from
Outfall, km 1 3 45 8 12
g C/m2 /growing season 174 132 96 51 48 30
mg C/m3 (max.)/day* 1092.6 950.8 533.9 133.9 99.6 58.9
*Average for the growing season (eleven measurements).
According to Jonasson (1969), in eutrophic waters phytoplankton primary production acts in the
following way.
Primary production (per cubic meter) increases very sharply
Fluctuations in primary production (per square meter) becomes irregular and
Annual primary production (per square meter) may increase sharply.
In coastal areas pulp mill effluents have a quite similar influence as in the lakes. Two series of
experiments of phytoplankton production in constant light, made in 1967 near the City of Kotka
in the Gulf of Finland, show this (Table 6). The phytoplankton primary production did increase to
a distance of 10-15 km from the outfall. In June, phytoplankton primary production did increase
to a distance of 10-15 km from the outfall. In June phytoplankton production was low and it did
not increase at 20 km distance from the outfall. In August eutrophication was obvious and it
reached beyond 20 km from the outfall.
CONCLUSIONS
Eutrophication of waters is a natural process, but as seen above, it can be greatly accelerated by
man. Therefore, it has become a recent problem in water protection. The increases in population,
industry and agricultural activities, for example, introduce excess nutrients and other pollutants
into the waters. Nutrients may accelerate the development of eutrophication.
-------
57
TABLE 6
Primary Production near the City of Kotka in the Gulf of Finland in 1967
Data from Lehmusluoto (1967)
Distance from
Outfall, km
June, mg C/m3/day
August, mg C/m3/day
2
14.1
15.8
6
56.7
63.6
10
78.5
115.9
15
59.8
219.3
20
39.7
175.9
6 C/M'/DAY
FIGURE 5 Phytoplankton primary production off the coast of Helsinki in the Gulf of Finland at
different stations on July 18,1967. The daily phytoplankton production is given in g
C/m2/day.
DISTANCE FROM OUTFALL KM
3-
2.
FMAMJ'J'A'SWN i I i i i i ' i i TTTT
MONTH
I 174 | I IK I
6 C/MVYEAR
Hoi
FIGURE 6 Phytoplankton primary production g C/m2/day (solid line) at different stations off
the coast of Helsinki in the Gulf of Finland in 1967. Total solar radiation (60° N) =
kcal/m2/min (clashed line) (Smayda, 1959). The annual phytoplankton production is
given in g C/m2.
-------
58
Phytoplankton primary production - as used as a measure of eutrophication - reflects the
ecological factors and algal succession. The most important factors are illumination, temperature
and nutrient concentration of the water. The annual phytoplankton production (intensity of
eutrophication) is almost wholly dependent on the nutrient concentration, and on the availability
of these nutrients. The influence of the other factors, despite annual fluctuations, may be almost
constant.
Domestic sewage and pulp mill effluents are important nutrient sources causing intense
eutrophication. Sewage is, thus, not the only type of waste water to cause eutrophication. Pulp
mill effluents may also play a great part in the process, although their first influence seems to be to
hinder phytoplankton production. At some distance from the outfall, when the effluents have
been diluted, phytoplankton production can proceed and the water eutrophicate.
The most important nutrients are nitrogen and phosphorus. Other substances, such as vitamines
and trace elements, must also be available. Many other growth stimulators are found in sewage.
It is important to consider all the possible efforts to reduce nutrient input to receiving waters in
any form in order to avoid overeutrophication of waters.
REFERENCES
Bringmann, G. and Kuhn, R. (1956) Der Algen-Titer als Massstab der Eutrophierung von Wasser
and Schlamm. Ges.-lng., 77, 374-381.
Demmerle, S. (1967) Der Einfluss von Humusstoffen auf das Algenwachstum, Manuscript.
Findenegg, I. (1964) BestimmungdesTrophiegradesvon Seen nach der Radiocarbon methode. Die
Naturwissenschaften, 15, 368-369.
Fonden, R. (1969) Heterotrophic bacteria in Lake Malaren and Lake Hjalmaren, Oikos, 20,
344-372.
Goldmann. C. R. (1962) A method of studying nutrient limiting factors in situ in water columns
isolated by polyethylene film, Limnol. Oceanogr., 7, 99-101.
Gran, H. H. (1929) Investigation of the production of plankton outside the Romsdalsfjord
1926-27, Rapp. Cons. Explor. Mer., 56, 1-112.
Hobbie, J. E. (1964) Carbon 14 measurements of primary production in two arctic Alaskan takes,
Verb. Internal. Verein. Limnol., 15,360-364.
Hubel, H. (1968) Die Bestimmung der Primarproduktion des Phytoplankton der Nord-Rugenschen
Boddengewasser unter Verwendung der Radiokohelnstoffmethode, Internal. Rev. ges.
HydrobJol., 53. 601-633.
-------
59
Jonasson, P. M. (1969) Bottom fauna and eutrophication. In Eutrophication: causes,
consequences, correctives, (ed. by National Academy of Sciences, Washington, D. C.)
274-305.
Jarnefelt, H. (1956) Zooplankton and Humuswasser, Ann. Acad. Sci. Fenn., A 31, 1-14.
Laaksonen, R. (1970) Vesistojen veden laatu, Vesiensuojelun valvontaviranomaisen vuosina
1962-1968 suorittamaan tarkkailuun perustuva tutkimus, Maa ja vesiteknillisia tutkimuksia,
17, 1-132, (English abstract).
Lehmusluoto, P. O. (1967) Selvitys kasviplanktonin perustuotannosta Kotkan edustan merialueella
vuonna 1967, Kymijoen vesiensuojeluyhdistys, 11, 1-7, (in Finnish).
Lehmusluoto, P. O. (1968) Kasviplanktonin perustuotanto Helsingin edustan merialueella,
Limnolgisymposion, 1967, 31-42, (English summary).
Lehmusluoto, P. 0. (1969) Veden pieneliotoiminnoista ja niiden mittaamisesta radioaktiivisen
hiilen avulla, Vesianalyyttisia menetelmia, Suomalaisten Kemistien Seura, 57-64, (in Finnish).
Lehmusluoto, P. O. and Heinonen, P. 0. (1970) Eraiden jatevesien vaikutus Saimaan
perustuotantoon, Vesi,4, 1-8, (in Finnish).
Leppanen, T. (1970) Tutkimuksia bakteerien gtukoosin kaytosta Hakojarvessa ja siihen
sijoitetuissa koealtaissa, Limnologian pro-gradu-tutkielma, 1-103, (in Finnish).
McCoy, E. I. and Sarles, W. B. (1969) Bacteria in lakes: populations and functional relations. In
Eutrophication: causes, consequences, correctives, (ed. by National Academy of Sciences,
Washington, D. C.), 331-339.
Ohle, W. (1940) Chemische Eigenart der smalandischen Seen, Verb. Internat. Verein. Limnol.,9,
145-159.
Ohle, W. (1956) Bioactivity, production, and energy utilization of lakes, Limnol. Oceanogr., 1,
139-149.
Pechlaner, R. (1964) Plankton production in natural lakes and hydroelectric water-basis in the
alpine region of the Austrian Alps, Verh. Internat. Verein. Limnol., 15,375-383.
Rodhe, W. (1958) Primarproduktion und Seetypen, Verh. Internat. Verein. Limnol., 13, 121-141.
Rodhe, W. (1969) Crystallization of eutrophication concepts in Northern Europe, In
Eutrophication: causes, consequences, correctives,(ed. by National Academy of QSciences,
Washington, D. C.), 50-64.
Rodhe. W.. Hobbie, J. E. and Wright, R. T. (1966) Phototrophy and heterotrophy in high
mountain lakes, Verh. Internat. Verein. Limnol., 16, 302-313.
Ryhanen, R. (1968) Die Bedeutung der Humussubstanzen im stoffhaushalt der Gewasser
Finnlands, Mitt. Internat. Verein. Limnol., 14, 168-178.
-------
60
Sederholm, H. (1963) Veden humus mikrobien ravintona, Limnologian pro-gradu-tutkielma, 1-65,
(in Finnish).
Siapiro, J. (1957) Chemical and biological studies on yellow organic acids of lake waters, Limnol.
Oceanogr.,2, 161-179.
Skulberg, O. (1964) Algal problems related to the eutrophication of European water supplies, and
a bio-assay method to assess fertilizing influences of pollution on inland waters. In Algae and
Man (ed. by D. Jackson, New York), 262-299.
Smayda, T. J. (1959) The seasonal incoming radiation in Norwegian and Arctic waters, and
indirect methods of measurement, J. Cons. Internal. Explor. Mer., 24, 215-220.
Steemann Nielsen, E. (1952) The use of radioactive carbon (C14) for measuring organic
production in the sea, J. Cons. Internal. Explor. Mer., 18, 117-140.
Thomas, A. E. (1968) Die Phosphattrophierung des Zurichsee und anderer Schweizer Seen., Mitt.
Internal. Verein. Limnol., 14, 231-242.
Wright. R. T. and Hobbie, J. E. (1965) The uptake of organic solutes in lake water, Limnol.
Oceanogr., 10,22-28.
-------
THE RECOVERY PROCESS OF A LAKE WHICH RECEIVED
WASTEWATER FROM AN ORE DRESSING PLANT
Bengt Ahling
A question, on which interest is currently focusing in the problem-complex of industrial pollution,
is what happens to a recipient of industrial waste water if the industry closes down or improves its
waste water plant so that the addition of polluted water appreciably diminishes.
The question is of very great importance from the standpoint to be adopted in reference to
measures for the restoration of polluted lakes. In those cases in which a recipient cannot, within a
reasonable time, purify itself to the point to which its water can be used for the different purposes
that may be considered desirable, it may be necessary to take steps to restore the lake.
Control lake
FIGURE 1 LakeBilsjan
6
-------
62
One of the lakes that has been investigated in order to study the recovery process is Lake Bal
(Balsjon) in Central Sweden. The area of Balsjon is 0.30 km2, and the mean and maximum depths
are 5 and 11 meters respectively. The catchment area is rather small, only 5.5 km2, and consists
mostly of marshes. Up to the autumn of 1967 Lake Bal was receiving the waste water from a
dressing plant for magnetite and hematite. The dressing plant operated with gravity and
wet-magnetic concentration and therefore had to use large quantities of water. As the smallest
particles were very hard to separate off with the methods used, they followed with the water
through the dressing plant.
The waste water was pumped to a banked-in area for sedimentation (Fig. 1). From here the still
turbid water was conducted via a settling dam to Balsjon. This made the water in Balsjon so turbid
that the Secchi disk transparency was reduced to only a few centimeters. Owing to this turbidity,
light could not penetrate to any appreciable depth, but was reflected or absorbed in the topmost
layer of water. In Balsjon there was constant sedimentation of the fine-grained particles. This
meant that any organic substance eventually formed in or added to the lake was immediately
embedded in the sediment, which made Balsjon a very sterile milieu for many organisms.
Before the plant closed down, Balsjon and a similar lake used as a control lake were investigated
for a couple of years, so that there was comparative material when it came to a study of the
recovery of the lake. In order to be able to give a clear picture of the recovery it may be as well to
give an account of the situation prevailing before the plant closed down.
QUALITY OF THE WATER WHEN THE DRESSING PLANT WAS OPERATING
Chemico-physical Conditions
Turbidity, Color and Secchi Disk Transparency
The most striking effect of the discharge from the dressing plant was the marked turbidity of the
recipient. The heamtite sludge that was not utilized imparted a distinct reddish color to Balsjon.
Through the discharge this coloring was much intensified. The control lake had a color value of
about 30 mg Pt/l, while values as high as 650 mg Pt/l were measured in Balsjon.
Owing to the turbidity, the Secchi disk transparency was reduced to only a few centimeters.
pHand Conductivity
The waste water had a pH between 8.2 and 8.3 in the sedimentation basins. In Balsjon, after
dilution, pH was between 7.8 and 8.1 in the spring and summer samples respectively. Owing to the
abundance of unbound ions in the waste water, the specific conductivity was considerably
increased. The value in Balsjon was about seven times that in the control lake.
Iron Content
Due to the waste sludge, the iron content of the water was for the most part very high, as the iron
content of the particles was about 15%. The highest content measured in Balsjon was 10 mg Fe/l,
which coincided with a markedly reddish coloring and turbidity. Analysis of the filtered and
unfiltered samples showed that most of the iron occurred as suspended material. A not
-------
63
inconsiderable part {25%), however, occurred in such a form that it passed a fine filter paper.
Content of Oxygen
The oxygen content was always high in Balsjon, showing values of about 90% saturation. The
waste from the dressing plant thus did not affect the oxygen content negatively.
Consumption of Permanganate
The consumption of potassium permanganate in the water was determined. The result showed that
very small amounts of organic chemically oxidizable substance existed in the water. The values
measured were about half of those for the control lake, or 10 mg KMn04/l.
Biological Conditions
The Balsjon water had a consistency of extremely fine-grained red sediment. This is unsuitable as a
substrate for more or less sessile living lake bed organisms. Balsjon was therefore characterized by
the absence of such species as well as the greater part of the submerged plants. In the samples
taken from the lake bed, only a few odd specimens were found of Chaoborus, which swims freely
in the water near the bed, and of crustaceans of copepod type. In the northern part of the lake
there were indications of organic flocculi, and in connection with these a few protozoons,
Flagellata, and diatoms like Synedra.
Owing to the marked turbidity in Balsjon, the light could not penetrate to any appreciable depth,
so that autotrophic organisms were able to exist only in a very thin surface layer. This implied a
marked diminution of the lake's total production.
Turbidity
ZP-units
19000
1100-
1000-
500-
400-
300-
200-
100
en
D
D White - Balsjbn
Black — Control lake
fll
—
1
-,
,
1967
68
69
year
FIGURE 2 Lake Balsjon: Turbidity vs. Time
-------
64
This meager production was especially noticeable in the result of the zooplankton counts, where
no individuals at all were found. There were, on the other hand, small specimens of pike (Esox
bicius), roach (Rutilus rutilus), perch (Perca fluviatilis) and freshwater crayfish (Potamobius
astacus).
THE RECOVERY PROCESS IN BALSJON AFTER THE CLOSING DOWN OF THE
DRESSING PLANT IN NOVEMBER 1967
Changes In The Physic-chemical Factors
Turbidity, Color, Iron Content and SecchiDisk Transparency
When the discharge of waste water stopped, the particles existing in the water settled. This resulted
in reduced turbidity (Fig. 2} and increased Secchi disk transparency (Fig. 3). The lake, which
earlier had had a marked red coloring, began to assume a considerably more normal hue. This is
also seen in the color measurements in Fig. 4, which show that the color values in Balsjon were
beginning more and more to agree with those in the control lake.
In connection with sedimentation of suspended particles, the iron content was of course reduced
to a range of values corresponding with that in the control lake (Fig. 5).
The variations now occurring in turbidity and in the Secchi disk transparency are due to the
addition of particles from the surrounding embankment of waste. After a spring flood or heavy
rainfall, this addition is noticed as a transient red coloring of Balsjon.
Secchi disk transparency
m
3-
White — Balsjon -
Black — Control lake
I
1
1967 68 69 70 71
FIGURE 3 Lake Balsjon: Secchi disk transparency vs. Time
year
-------
65
Color 4OO
mg pt /1
250
200-
150-
100-
50-
25-
White - Balsjon
Black - Control lake
•
n
1967 68 69 70 71
FIGURE 4 Lake Balsjon: Color vs. Time
Total iron
mg Fe/ I 9 70
year
25-
2.0-
1.5-
1.0-
0.5-
u
1
i
"
,
White - Balsjdn
Black - Control It
I 1 Ifin
1967 68 69 70 71 year
FIGURE 5 Lake Balsjon: Iron content vs. Time
pHand Conductivity
A certain reduction of pH was observable in the spring samples, while the autumn samples even
showed an increase. This may be interpreted to mean that the concentration of the pH-increasing
substances had been reduced, which led in turn to a reduction of pH. Furthermore, the primary
production increased, which gave increased pH during the production period.
The big difference earlier noted between the specific conductivity of Bilsjon and the control lake
respectively began gradually to lessen. (Fig. 6). The slowness of this lessening was due to the very
slight water turnover in the lake.
-------
66
Electrical conductivity
Of 2
300-
200-
100-
0-10"
White - Batsjbn
1
1
Black - Control lake
1
n
n
i in
i
1967
69
70
year
FIGURE 6 Lake Balsjon: Electrical conductivity vs. Time
Oxygen Content and Consumption of Permanganate
After the closing down of the dressing plant the values for the oxygen content were still high. In
some samples of water from the bottom of the lake, however, an oxygen deficit down to 17%
saturation was observable. This indicates that organic substance may have been added or formed in
the lake in such quantities as to have had an effect on the oxygen economy of the lake.
The consumption of permanganate had increased to such an extent in Balsjon that it coincided
with that in the control lake, (Fig. 7) which implied that the quantity of organic substances in the
Consumption of
Permanganate
mg KMnO^ White — Balsjbn
70-
60-
50 •
40
30 •
20 -
10
1 n
11 II
Black — Control lake
''1
II. 1
1967 68 69 70 71 year
FIGURE 7 Lake Balsjon: Permanganate consumption vs. Time
-------
67
water had increased. This increase was an indication that the primary production in the lake had
increased, or that permanganate-consuming substance had been added to the lake.
Nitrogen and Phosphorus
Nitrogen and phosphorus show higher values as in the control lake. There has not yet, however,
been any change in the values.
Changes In The Biological Conditions
Phytoplankton
After the disappearance of turbidity the occurrence of phytoplankton markedly increased (Fig. 8).
This applied especially to the diatoms, where during the autumn and winter Cyclotella comensis
and to a certain extent also Synedra acus showed high values for the number of individuals (on 4
November 1969, for instance, there were 3 million Cyclotella per liter). In the spring and summer
the Chrysomonad Rhodomonas lacustris and the earlier completely absent Dinobryon divergens
occurred in considerable quantities.
Apart from this increased number of individuals, the number of species, especially of Chlorophyta,
increased (Fig. 9). Newly added species were e.g. Elakatothrix gelatinosa and Gloeocystis. The new
species in Balsjon were still of very slight importance for the total production of the lake. From
the general picture of plankton it emerged that there was a relatively large' number of individuals
belonging to a small number of species (Fig. 10); this gave the picture of an extreme milieu.
However, the increasing number was an indication that the milieu was becoming less extreme.
If we compare the composition of species in Balsjon with that in the control lake (Fig. 11) we find
a considerably greater variety in the latter, with a larger number of species represented.
3-1
!05-
Total numbers
of individual
phytoptankton
White — Balsjon
Black - Control lake
67
68
59
70
71
year
FIGURES Lake Balsjon: Phytoplankton vs. Time
-------
68
30-
20-
10-
Number of species of
phytoplankton
I
I
White — Balsjcn
Black -Control lake
1967
68
69
70
year
FIGURE 9 Lake Balsjon: Phytoptankton species vs. Time
Zooplankton
Two years after the closing down of the dressing plant Balsjon showed a zooplankton composition
almost identical with that in the control lake. The differences referred exclusively to the number
of individuals. Thus the number of Ciliata in the control lake was about 10 times greater than in
Balsjon, and other zooplankton were 5 times more numerous. Although Balsjon still showed a
paucity in the number of individuals as compared with the control lake, the zooplankton existing
there implied a marked increase compared with the earlier total absence of these organisms.
Among the zooplankton occurring, the predominant species were the rotifers Polyarthra,
Gastropus stylifer and Keratella cochlearis. Other zooplankton species of Cladocora and copepoda
occurred in Balsjon only as isolated individuals.
Lake Bed Fauna
When the rain of particles over the sediment surface stopped, the fauna normally living on the lake
bed could once more start colonizing it. During the investigations carried out the year after the
closing down of the dressing plant, it was possible to observe only very slight changes. One only
found scattered specimens of Ephemeridae and Chironomidae in the lake bed samples taken.
It was not until one year later, during the summer and autumn of 1969, that it was possible to
observe that a community of organisms was in the process of being built up. Ephemeridae, in
1970, were of general occurrence. It was, however, only Ephemera vulgata that occurred, and this
gave an impression of the instability of the system.
The groups of lake bed organisms most common in Balsjon two years after the closing down of the
dressing plant were Chironomidae and Ceratopogonidae. Of the Chironomidae, Chironomus
plumosus were abundantly represented, but large numbers of organisms from the subfamily
-------
Balsjbn
100-
90-
80-
:
:
- -
40-
:..
;; .
.
'A
V,
Y;
X
— 1 fr
y
/
y
y
I
_
y
^
^
y
y
^
,
'
• -
1967
:
-
69
• Cyanophyta
^ Chlorophyta
H Chrysomonadinae
^ Diatomeae
C Pyrrophyta
71
year
FIGURE 10 Lake Balsjon: Percent of species vs. Time
.
• Cycnophytc
E Chlorophyta
H Chrysomonadinae
0 Diatomeae
LH Pyrrophyta
Control lake
100-
90-
:
-
:
40
30
.
7
^
2.
\
1967 68 69 70 year
FIGURE 11 Control Lake: Percent of species vs. Time
Tanypodinae also occurred. Of the Ceratopogonidae, Palpomyia was the animal occurring most
generally in the whole system even if a smaller number of Chaoborus also occurred. It was possibl
to find a greater density of organisms if one went to the northern part of the lake. This was the
part furthest from the region where the waste was introduced into Balsjon, so it thus got the I
sedimentation of particles over the sediment surface. It was also possible to find a considerably
more modified community of organisms in the discharge canal in the south end of the lake
the water had little depth and the surrounding reeds contributed organic material which promoted
-------
70
the colonization of the lake bed organisms. Apart from the already mentioned organisms one
found in this canal Herpobdella, Asellus and large quantities of mites.
Owing to the small amount of organic material on the bottom, the lake bed organisms had very
poor protection against predacity. This predacity must have certainly been a strongly reducing
factor as regards these organisms. Analysis of the gastric contents of fish from Balsjon also showed
the organisms mentioned in the foregoing.
Other Changes
Apart from the changes mentioned in the community of organisms, it was possible to make a
number of studies through visual examination of the lake. These changes were very difficult to
assess quantitatively, and the observations thus made were, accordingly, very subjective.
On the lake shores one could see great swarms of the ostracod Notodromas monacha. This
organism commonly occurs in stagnant little lakes during the summer. It is, however, extremely
unusual to find such mass development as was here.
At the edge of the shore at the populations of Phragmites that existed, it was possible to observe
large shoals of one-year-old perch and a large number of pike. Such shoals often occurred near
shoals of Notodromas, which indicates that the ostracod may be an object of nourishment for the
small fry among the fish.
The first traces of Periphyton began to appear on the lake bed. These were the thread-formed algae
Zygnema and Spirogyra which had begun to spread to the regions where the light was earlier too
weak for these autotrophic algae.
In the cores of sediment collected in Balsjon in 1969 it was possible to find traces of organic
material. This was composed of detritus and vegetable fiber that was presumably only partly
produced in the lake and had otherwise come from the soil surrounding the lake. Besides this
material there were also shells and remains of diatoms and Crustacea.
This study of lake recovery reached one of the most interesting stages in 1970. The first part of
the recovery referred chiefly to the physio-chemical factors and was a relatively simple process
which was mainly a question of sedimentation rates. The most important question for this process
in the near future is how rapidly the embankments of waste along the shores can be bound by
vegetation. This is important for the prevention of inorganic material being washed out into the
lake by rain and melt water.
The second stage, which referred chiefly to the building up of a balanced biosystem, is a good deal
more difficult to foretell. The results hitherto obtained show that the number of species in the
system is on the increase. They also show that certain species very easily begin to become
predominant with a mass development. This is presumably a tendency that will continue until an
organic layer has been formed above the inorganic sediment. This organic layer is essential to give
the different organisms protection against predacity. When they have this protection the
fluctuations due to this predacity will be reduced.
-------
DEPLETION OF OXYGEN BY MICROORGANISMS
IN ALASKAN RIVERS AT LOW TEMPERATURES
Ronald C. Gordon
INTRODUCTION
Several arctic (Lotspeich, unpubl; Schallock, unpubl) and subarctic (Prey, 1969; Gordon, unpubl;
Mueller, unpubl; Schallock, unpubl) rivers in Alaska (Alaskan rivers) have low concentrations of
dissolved oxygen (DO) during periods of total ice cover; conditions which occur naturally without
domestic or industrial pollution. A similar oxygen deficit was noted in some unpolluted rivers in
the northern and central belt of the U.S.S.R. (Drachev, 1964). Data from various subarctic rivers
in Alaska indicated that DO depletion was a continuous process throughout most of the period of
ice cover, with an increase in DO concentration shortly before spring breakup (Frey, 1969;
Gordon, unpubl; Schallock, unpubl). The extent of depletion increased progressively toward the
lower reaches of each river (Frey, unpubl; Gordon, unpubl).
Investigations in the U.S.S.R. have shown that the low DO concentration resulted from the ice
cover which prevented reaeration (Drachev, 1964). Since there is essentially no open water during
the period of ice cover over many Alaskan rivers, there is little chance for significant reaeration.
Under natural conditions, the extent of oxygen depletion is often sufficient to reduce the DO
concentration to a level far below the 7 mg/l minimum set by the Alaska water quality standards
(State of Alaska, 1967). A DO concentration of 1.1 mg/l was measured in an unpolluted arctic
river {Lotspeich, unpubl; Schallock, unpubl) and, 1.1 mg/l (Gordon, unpubl) and 1.0 mg/l
(Roguski, 1967) in unpolluted subarctic rivers.
The aquatic biota of Alaskan rivers seem to survive the extreme fluctuations in the amount of DO
which they encounter throughout the year under natural conditions. Problems arise when
oxidizable domestic or industrial wastes enter these rivers. When the biochemical oxygen demand
(BOD) of these wastes is added to the natural requirement for DO, the result may be detrimental
to the ecosystem.
Ingraham and Stokes (1959) discussed the numerous definitions of psychrophilic bacteria and set
forth what is probably the most useful definition, "Psychrophiles are bacteria that grow well at 0°
C within 2 weeks". These organisms appear to be ubiquitous in nature since they have been found
in soil, rivers, lakes, mud and food (Farrel and Rose, 1967; Stefaniak, 1968, Stokes and Redmond,
1966). Psychrophiles have been studied in both the Arctic and Antarctic and have been found in
soil and water (Boyd, 1958; Boyd and Boyd, 1967; Fournelle, 1967; McDonald, et al., 1963;
Straka and Stokes, 1960). These organisms and their activity at low temperatures have been the
subject of several reviews (Farrell and Rose, 1965; 1967; Ingraham and Maaloe, 1967; Ingraham
and Stokes, 1959; Miller, 1967) and will not be discussed in detail here. Stokes and Redmond
(1966) considered psychrophiles to be present in large enough numbers in natural habitats to be
important in the cycling of matter. Wuhrmann, et al. (1966) stated, "Self-purification processes
start at the microbial level. . ." and, "Most of the work is accomplished by heterotrophic
microorganisms (bacteria, fungi, flagellates)".
71
-------
72
Active metabolism of organic material in a river during the winter has been demonstrated in the
U.S.S.R. (Drachev, 1964). Plate counts of heterotrophic bacteria indicated that Alaskan rivers have
bacterial populations in the range of 104 - 106 organisms/ml which are capable of growth on a
synthetic medium at low temperatures(Gordon, unpubl). There is evidence that the number of
organisms capable of growth at low temperatures increases progressively toward the lower reaches
of a subarctic river in Alaska (Gordon, unpubl) and in a river in the U.S.S.R. (Drachev, 1964)
during the period of total ice cover.
It has been shown that psychrophilic bacteria are capable of rapid metabolic activity at low
temperatures. Since Alaskan rivers have populations of these organisms, it appears that they may
be responsible for a significant portion of the DO depletion observed under both natural and
polluted conditions. The subject of this report is DO depletion by the indigenous bacteria in a
subarctic river. The effect of added organic and inorganic nutrients and incubation temperature on
the rate and extent of DO depletion was investigated. The data obtained from a subarctic river
were compared to similar data from an arctic river.
MATERIALS AND METHODS
River Description And Sampling Locations
Most of the experimental results were obtained from a subarctic river in interior Alaska. Because
of the high level of domestic pollution in the lower reach and the convenient location, the Chena
River was chosen for detailed study. It is a nonglacial stream with many ground water sources, and
is approximately 150 miles in length (Frey, 1969). Raw domestic sewage and effluents from
several primary sewage treatment plants in the greater Fairbanks area enter the river in the last 28
miles before it joins the Tanana River. Two sampling locations were selected, one below all major
sources of domestic pollution and the other above any source.
Comparative data were obtained from an arctic river in the "Arctic Slope" area of Alaska. The
Sagavanirktok (Sag) River was selected because it is the major river flowing through an area of
extensive oil development and is accessible for sampling. It is a nonglacial stream originating in the
Brooks Range, flowing north approximately 170 miles to the Beaufort Sea and receiving little, if
any, domestic pollution (Gordon, unpubl). One sampling location was selected approximately 85
river miles above the mouth of the river near the settlement of Sagwon.
Sample Collection And Handling
All sample locations had total ice cover and a water temperature of essentially 0° C throughout the
study period. Samples were obtained through holes drilled in the ice. Samples from the Chena
River for the dissolved oxygen (DO) depletion study were collected in sterile five gallon
polypropylene carboys by dipping water from the hole in the ice. Because of the large volume
required, no problem with increase of water temperature was encountered during the 2 to 3 hour
period between sample collection and handling in the laboratory.
Samples from the Sag River for the DO depletion study required somewhat different handling.
These samples were collected in new, clean, but not sterile, five gallon polyethylene carboys which
were sealed with tight screwcaps. It was not possible to dip the water, so it was pumped into the
carboys. The samples were shipped by air freight to the laboratory. The water temperature rose
-------
73
from approximately 0° to 3.5° C during the 9 hour period between sample collection and handling
in the laboratory. This temperature rise did not appear to be excessive, and was not considered
significant.
Samples for chemical analysis were collected in small mouth 250 ml screwcap polyethylene
bottles. These bottles were filled by submerging them in the hole drilled in the ice and returned to
the laboratory without further field treatment. They were frozen as soon as possible after arrival at
the laboratory and stored at -20° C until they were analyzed.
Samples for the determination of DO in the river were collected in 300 ml biochemical oxygen
demand (BOD) bottles. The bottles were lowered on a rod sampler below the bottom of the ice,
allowed to fill completely, and the oxygen was fixed immediately after being brought to the
surface.
Handling Of Samples In The Laboratory For DO Depletion Studies
After the samples were returned to the laboratory, they were taken directly into the 10° C cold
room. The DO depletion study was set up immediately, using pre-cooled glassware to minimize
any adverse effect on the natural distribution of the microorganisms in the samples. A
predetermined volume of river water, 24-36 liters, and the substrate being studied were placed in a
2'/2 or 3'/2 gallon, sterile, glass carboy. The carboy was placed on the apparatus shown in Figure
1-A. The water was stirred rapidly with a magnetic stirrer while the temperature was raised 1° -
FIGURE 1 Apparatus for preparation and bottling of river water samples for dissolved oxygen
depletion studies. (A) Sample was stirred vigorously while the temperature was
equilibrated at 1° - 1.5° C above the desired incubation temperature with a
thermostatically controlled, 1000 watt, Vicor glass, immersion heater; followed by
dissolved oxygen equilibration at or near saturation by aeration with a gas dispersion
tube. (6) Equilibrated sample was pumped into biochemical oxygen demand bottles
for incubation.
-------
74
1.5° C above the intended incubation temperature with a thermostatically controlled, 1000 watt,
Vicor glass, immersion heater. The increase in temperature above that selected for incubation
prevented supersaturation of the water with DO. When the desired temperature was reached,
stirring was continued and the water was aerated vigorously for 10 minutes using a gas dispersion
tube to bring the DO level to or near saturation. After temperature adjustment and aeration, the
river water was pumped into BOD bottles as shown in Figure 1-B. The bottles were filled from the
bottom to prevent entrainment of additional DO. The initial DO level was determined by
immediately fixing the oxygen in three of the BOD bottles. The rest of the BOD bottles were
placed in incubators at 0°, 5°, 10°, 15° or 20° C. Time intervals were selected to permit the
depletion of DO to be followed. The DO was determined in three bottles at each time interval.
Substrates Used For DO Depletion Studies
Several laboratory substrates of varying complexity were used. Vitamin-Free Casamino Acids,
Control 534363 (Difco) were used to compare rates of DO depletion at several temperatures, with
water from various sources, and as a control for other studies. Yeast Extract, Control 523143
(Difco) and Beef Extract, Control 495576 (Difco) were used as complex substrates containing
growth factors. Growth factors are defined as organic compounds, generally in minute amounts,
required for growth by an organism in addition to the principal sources of carbon and energy.
Glucose (Dextrose, Control 527712, Difco) was used to represent the carbohydrates. Ethyl alcohol
(dehydrated, N.F., Federal Government stock no. 6505-105-0000) was the only alcohol used.
Sodium acetate NaC2H3O2-3H2O, Mallinckrodt analytical reagent) was used to represent the
organic acids.
Primary and secondary sewage treatment plant effluents were also studied. Primary effluent was
obtained from the Fairbanks city plant before the effluent entered the chlorine contact chamber.
Secondary effluent was obtained from a bench scale activated sludge system being operated in the
Alaska Water Laboratory at 0° - 1.0° C.
Enumeration And Isolation Of Heterotrophic Bacteria
The membrane filter method and a broth culture medium prepared from components [2.5 g/l
Yeast Extract (Difco), 5 g/I Tryptone (Difco), and 1 g/l Dextrose (Difco) made up in glass distilled
water and adjusted to pH 7.0 at 25° C before autoclaving] were used to enumerate bacteria at 0°
C. This medium was found to give higher numbers on membrane filters at 0° C than any other
medium tried (Gordon, unpubl). However, this does not mean that these were the only bacteria
present in the water. All membrane filter preparation was done in the 10° C cold room, using
pre-cooled equipment and materials. Incubation of filters was continued until there was no further
increase in numbers on consecutive counts.
Isolation of pure cultures was accomplished by picking individual colonies from the membrane
filter after the number of colonies had stopped increasing. The colonies were placed in tubes of the
same broth medium used for initial enumeration, and incubated at 5° C because growth was more
rapid than at 0° C. After growth appeared, material from the broth cultures was streaked on Plate
Count Agar (Difco) and incubated at 5° C. Individual colonies were picked and grown in broth.
This procedure was repeated as a final check of culture purity. The pure cultures were maintained
for further study by monthly transfer to fresh broth and incubation at 5° C.
-------
75
Chemical Analyses
The Technicon Auto Analyzer was used for the following analyses: Orthophosphate phosphorus
by trie Technicon ammonium molybdate industrial method; ammonia nitrogen by the sodium
phenolate method (FWQA, 1969); nitrite nitrogen by diazotization (FWQA, 1969); nitrate
nitrogen by hydrazine reduction (FWQA, 1969).
Total nitrogen and total carbon were determined with the Perkin-Elmer model 240 Elemental
Analyzer.
Total phosphorus was determined by the persulfate digestion method (FWQA, 1969), except for
glucose. Glucose samples were ashed (AOAC, 1965), followed by the orthophosphate phosphorus
determination previously described.
Chemical oxygen demand was determined as described in the 12th edition of Standard Methods
for the Examination of Water and Wastewater (APHA, 1965).
DO was determined by the azide modification of the iodometric method (APHA, 1965).
Statistical Treatment Of DO Depletion Data
Each set of 3 DO measurements was evaluated by the Q Test to reject questionable results. The
remaining measurements were averaged to obtain the reported result. To compare rates of DO
depletion, an attempt was made to establish a rate constant with one substrate at each incubation
temperature. Data obtained during the period of most rapid DO depletion were treated with first
and second order kinetics, and did not fit either form. The arithmetic form, DO vs time, provided
the most useful treatment of the data. A straight edge was laid along the slope of the DO depletion
curve, and the data points on the portion of the curve which appeared to have the most rapid rate
of change were used to establish an approximate rate (mg/l/hr) for the purpose of comparing data
within this study.
RESULTS
Pure Culture Study Of Psychrophilic Bacteria Isolated From A Subarctic River
Water samples were collected from a polluted and an unpolluted location in a subarctic (Chena)
river on December 17. 1968 and contained, respectively, 9000 and 550 heterotrophic bacteria per
ml which were capable of growth at 0° C on the complex organic medium as described in the
Materials and Methods section. AH colonies on a representative membrane filter from each location
were isolated in pure culture. Broth tubes inoculated with the pure cultures were incubated as
shown in Table 1. All cultures from both locations grew at 0°, 5°, and 10° C, but not at higher
temperatures. The percentage of the total number which did grow at 20° C and 25° C was the
same from both locations. At 30° C and above, the percentage of cultures from the polluted
location that grew decreased much more slowly than those from the unpolluted location. This
suggested that domestic pollution caused a change in population composition.
-------
76
Parrel I and Rose (1967) pointed out in their review that Gram negative rods are the most common
psychrophilic bacteria isolated, both qualitatively and quantitatively. Gram negative rods have
been isolated from littoral and marine sediments in the Canadian Arctic (McDonald, et al., 1963)
and were the most common bacteria isolated from water in subarctic Alaska (Fournelle, 1967).
Farrell and Rose (1967) referred to phychrophilic members of the genus Vibrio (spiral bacteria) as
not being as common as the Gram negative rods, but still isolated regularly.
Further study of the pure cultures revealed that only Gram negative rod and spiral morphological
types of bacteria had produced colonies on the original membrane filters. The effects of
incubation temperatures on the two types of bacteria from each location are shown in Table 2.
The data, from the unpolluted location, indicated that increasing the incubation temperature
above 25° C caused a more rapid decrease in the percentage of the spiral than of the rod shaped
bacteria which grew. The results from the polluted location were similar except that the more
rapid decrease of spiral bacteria took place above 30° C rather than 25° C. An additional point of
interest was that all the spiral bacteria grew at 20° C, but some of the rods from both locations
were inhibited at this temperature.
Examination of Table 2 shows that the ratio of rod to spiral bacteria changed from 1.5:1 at the
unpolluted location to 2.9:1 at the polluted location. This twofold increase of rods relative to
spiral bacteria was further indication that domestic pollution altered the composition of the
bacterial population.
TABLE 1
Effect of Increased Incubation Temperature on the Growth of Bacterial Isolates
from Samples Obtained from a Subarctic River3'
Incubation
Temperature _ Sample Location H
Polluted? Unpolluted"
Number of % of Total Number of % of Total
Isolates Isolates Isolates Isolates
0° -10° C 66 100 38 100
20 63 95.5 35 92.1
25 52 78.8 30 78.9
30 49 74.2 11 28.9
35 22 33.3 6 15.8
45 7 10.6 1 2.6
Total 66 38
a. Samples were taken on December 17, 1968, when the river had total ice cover and the
water temperature was 0 C.
b. The isolates were obtained by picking all colonies from a membrane filter which had
been incubated at 0° C until there was no further increase in numbers on consecutive
counts.
c. The polluted location was below a reach of the river receiving raw domestic sewage and
effluents from primary sewage treatment plants.
d. The unpolluted location was upstream from any source of domestic or industrial
pollution.
-------
77
TABLE 2
Relative Distribution and the Effects of Increased Incubation
Temperature on the Growth of Two Morphological Types of Bacteria
Isolated from a Subarctic River3'
Incubation
Temperature
0-10 C
20
25
30
35
45
Total
Rod
Morphological Type
Spiral
Sample Location
Polluted^ Unpolluted?
Sample Location
Polluted
No. of
Isolates
49
46
46
35
19
7
%of
Isolates
100
93.9
93.9
71.4
38.8
14.3
No. of
Isolates
22
20
18
10
5
1
%of
Isolates
100
90.9
81.8
45.5
22.7
4.5
No. of
Isolates
17
17
16
14
3
0
%of
Isolates
100
100
94.1
82.4
17.6
0.0
No. of
Isolates
15
15
12
1
1
0
%of
Isolates
100
100
80.0
6.7
6.7
0.0
49
22
17
15
a. Samples were taken on December 17, 1968, when the river had total ice cover and the
water temperature was 0 C.
b. The isolates were obtained by picking all colonies from a membrane filter which had
been incubated at 0 C until there was no further increase in numbers on consecutive
counts.
°' ^ P°lluted location was below a reach of the river receiving raw domestic sewage and
effluents from primary sewage treatment plants.
d. The unpolluted location was upstream from any source of domestic or industrial
pollution.
Effect Of Complex Organic Substrate Concentration And Incubation Temperature On The
Dissolved Oxygen (DO) Depletion In Subarctic River Water
Vitamin-free casamino acids had been used previously in pure culture studies of Pseudomonas
fluorescens at low incubation temperatures (Jezeski and Olsen, 1961; Olsen and Jezeski, 1963) and
were found to give excellent growth, which was not enhanced by the addition of yeast extract. In
view of these earlier reports, some preliminary results from this laboratory and the relatively
simple composition of the substrate, vitamin-free casamino acids were selected as the baseline and
comparative substrate.
The effect of substrate concentration on DO depletion in polluted river water is shown in Figure 2.
These data indicated that a vitamin-free casamino acids concentration of 120 mg/l was sufficient
to eliminate the substrate as a rate limiting factor in DO depletion. There was no lag phase at 20°
C and the DO concentration in the water was reduced to nearly 0 mg/l in 15-16 hours. A similar
effect of substrate concentration was observed at 10° C, but the time required to deplete the DO
from near saturation to 0 mg/l was approximately 50 hours.
-------
78
68 10 12 W
TME (HOURS)
FIGURE 2 Effect of the concentration of a complex organic substrate on dissolved oxygen
depletion in sub-Arctic river water polluted with raw domestic sewage and effluents
from primary treatment plants. Samples were incubated at 20° C. Symbols: •, 90
mg/l; A, 120 mg/l; and A, 150 mg/l vitamin-free casamino acids; o, river water blank.
Since the water temperature in the Chena River rarely, if ever, rises above 20° C (Frey, 1969;
Gordon, unpubl), temperatures between 0° and 20° C were selected for incubating samples. The
effect of incubation temperature on DO depletion in polluted Chena River water is shown in
Figure 3. The volume of water obtained from the river was large enough to supply samples for all
incubation temperatures. This provided directly comparable temperature effect data when the
samples were incubated in the presence of 120 mg/l vitamin-free casamino acids. The results
indicated that the length of the acceleration phase increased and the rate of DO depletion was
reduced as the incubation temperature was decreased, and there was a short lag phase at the 0° C
incubation temperature. However, the extent of DO depletion did not appear to be temperature
dependent.
Comparative results on the effect of incubation temperature were obtained with unpolluted river
water (Fig. 4). The results showed that there was a lag phase at the lower incubation temperatures
(0°, 5°, and 10° C) before the acceleration phase began. This was in contrast to the lack of a lag
phase with samples from the polluted location. The extent of the DO depletion, as found with the
sample from the polluted location, did not appear to be temperature dependent. However, the
total elapsed time was increased 50-100 percent.
The Relative Effect Of Complex Organic Substrates On DO Depletion In Subarctic River Water
Most of the psychrophilic bacteria are found in a few genera (Farrell and Rose, 1967) and some
have been isolated from water in subarctic Alaska (Fournelle, 1967). Nutritional studies.have
shown that the growth requirements vary over a wide range, from the simple need for a carbon and
energy source to the need for vitamins and other preformed growth factors (Adams and Stokes,
-------
79
2 -
SO 100 120
TIME (HOURS)
200
FIGURE 3 Effect of incubation temperature on dissolved oxygen depletion when 120 mg/l
vitamin-free casamino acids were added to sub-Arctic river water polluted with raw
domestic sewage and effluents from primary sewage treatment plants. A river water
blank (A) was incubated at 20° C.
1968; Jezeski and Olsen, 1961; Mulder, 1964, Olsen and Jezeski, 1963; Pereira and Morgan, 1957;
Prince, et al., 1954). Vitamin-free casamino acids contain 18 amino acids and essentially no other
growth factors, while yeast and beef extracts contain many amino acids, vitamins and other water
soluble growth factors. The use of these three substrates for DO depletion studies permitted an
examination of the effect of added growth factors.
40
80 120 160 200 240 280
TIME (HOURS)
FIGURE 4 Effect of incubation temperature on dissolved oxygen depletion when 120 mg/l
vitamin-free casamino acids were added to unpolluted sub-Arctic river water. A river
water blank (A) was incubated at 20° C.
-------
80
All the bacteria isolated from both locations were capable of growth at 10° C on a complex
medium containing a variety of preformed growth factors (Table 1), and DO depletion with river
water samples took place in a reasonable time at 10° C with vitamin-free casamino acids as the
substrate (Figs. 3 and 4). Since 10° C appeared to be adequate for growth and metabolic activity,
it was selected as the incubation temperature for additional studies.
Water samples from both sample locations were incubated at 10° C with the three complex organic
substrates in quantities containing the same amounts of carbon. The results are presented in
Figures 5 and 6. The acceleration phase of the DO depletion curve was shorter with samples from
both locations when the yeast or beef extract was used as the substrate. This suggested that
preformed growth factors either enhanced overall metabolic activity or were required by a portion
of the bacterial population. The results indicated that there was a difference in the relative effect
of yeast and beef extracts on DO depletion at each location. The yeast extract caused a very
pronounced decrease in the acceleration phase as related to either of the other substrates when
incubated with water from the unpolluted location (Fig. 6), while the effect in water from the
polluted location did not become apparent until later (Fig. 5). This could mean (a) that one or
more growth factors were added with the sewage or (b) that the bacteria enhanced by domestic
pollution (Table 1) did not require the growth factors in yeast extract and that those which
required growth factors needed a much longer time to utilize significant DO.
20 30 40
TIME (HOURS)
50
FIGURE 5 Relative effect of three complex organic substrates on dissolved oxygen depletion in
sub-Arctic river water polluted with raw domestic sewage and effluents from primary
treatment plants. Samples were incubated at 10° C. Symbols: o, river water blank; A,
120 mg/l vitamin-free casamino acids; •, 106 mg/l beef extract; O, 80 mg/l yeast
extract. All three substrates contained equal amounts of carbon.
-------
81
30 40 5O
TIME (HOURS)
FIGURE 6 Relative effect of high levels (substrate not rate limiting) of three complex organic
substrates on dissolved oxygen depletion in unpolluted sub-Arctic river water when
incubated at 10° C. Symbols: o, river water blank; •, 120 mg/l vitamin-free casamino
acids; A, 106 mg/l beef extract; A, 80 mg/l yeast extract. All three substrates
contained equal amounts of carbon.
Results similar to those obtained with a high level of substrates in unpolluted water (Fig. 6} were
obtained with a low substrate level, shown in Figure 7. In all cases, this low level of substrate
limited the amount of DO utilized. Growth factors added in the yeast and beef extracts shortened
the acceleration phase, but the extent of DO utilization with these substrates was less than with
the vitamin-free casamino acids. This suggested that one or more amino acids were required by a
large portion of the bacterial population and that there was a limiting amount present in the
extracts. Similar results were obtained at 0°, 5°, 15° and 20° C with this low substrate level. Since
these data would be redundant, they have not been shown.
The Effect On DO Depletion When Nitrogen And Phosphorus Were Added To Subarctic River
Water In The Presence Of Substrates Devoid Of These Nutrients
Ammonia, nitrite and nitrate nitrogen and orthophosphate phosphorus concentrations were
determined by chemical analysis each time samples were taken from either location, and the ranges
of values obtained are shown in Table 3. Both ammonia nitrogen and orthophosphate phosphorus
were increased by domestic pollution.
The results in Figures 8 and 9 indicated that glucose, which contained the same amount of carbon
as the vitamin-free casamino acids control, was poorly utilized as a substrate for DO depletion in
Chena River water. Nitrogen and phosphorus, in amounts equal to the amounts in vitamin-free
-------
82
casamino acids, were added to the river water, which contained glucose. When nitrogen alone was
added to the system, little effect on the DO depletion was observed with either polluted (Fig. 8} or
unpolluted (Fig. 9) water. The same was true for phosphorus in the polluted water. However,
when phosphorus was added to the unpolluted water, DO depletion appeared to be enhanced to
some extent. This suggested that the amount of phosphorus naturally present was a limiting factor.
When phosphorus and nitrogen were both added, a very marked effect on DO depletion in the
40 60 6O
TIME (HOURS)
FIGURE?
Relative effect of low levels (substrate being rate limiting) of three complex organic
substrates on dissolved oxygen depletion in unpolluted sub-Arctic river water when
incubated at 10° C. Symbols: o, river water blank; •, 30 mg/l vitamin-free casamino
acids; D, 26 mg/l beef extract; •, 20 mg/l yeast extract. All three substrates
contained equal amounts of carbon.
TABLE 3
Chemical Analysis of Water Samples from Two
Locations on a Subarctic River
Determination
Dissolved Oxygen
Ammonia Nitrogen
Nitrate Nitrogen
Nitrite Nitrogen
Orthosphate
Phosphorus
Polluted Sample
Location3
2.5 - 5.9
0.40 -0.83
0.02 -0.09
0.003 - 0.007
0.02 -0.08
Range of Values (mg/l/hr)
Unpolluted Sample
Location11
3.5 -8.0
0.06 -0.18
0.03 -0.12
0.001 - 0.004
<0.01 - 0.02
a. The polluted location was below a reach of the river receiving raw domestic sewage and
effluents from primary sewage treatment plants. The range of values is from 10 samples
taken between December 16,1969 and April 7,1970.
b. The unpolluted location was upstream from any source of domestic or industrial
pollution. The range of values is from 7 samples taken between December 10,1969 and
April 14,1970.
-------
83
40 50 60
TIME (HOURS)
70 80 90 100
FIGURE 8
Effect of glucose on dissolved oxygen depletion in sub-Arctic river water polluted
with raw domestic sewage and effluents from primary treatment plants. Incubation
at 10° C in the presence and absence of added inorganic nitrogen and phosphorus.
Symbols: o, river water blank; A, 120 mg/l vitamin-free casamino acids as a control;
•, 80 mg/l glucose; D, 80 mg/l glucose, KH2PO4 (0.33 mg/l phosphorus) and
K2HPO4 (0.33 mg/l phosphorus); •, 80 mg/l glucose. (NH4)2SO4 (3.33 mg/l
nitrogen) and KNO3 (10 mg/l nitrogen); A, 80 mg. I glucose, K2HPO4, KH2P04,
(NH4)2SO4 and KNO3 (nitrogen and phosphorus in same amounts as above). The
glucose, K2HPO4f KH2PO4, (NH4)2SO4 and KN03 were added to give the same
levels of carbon, phosphorus, and nitrogen as found in the casamino acids control.
-------
84
20 40
60
80 100 120
TIME (HOURS)
140 160 180 200
FIGURE 9 Effect of glucose on dissolved oxygen depletion in unpolluted sub-Arctic river water
when incubated at 10° C in the presence and absence of added inorganic nitrogen
and phosphorus. Symbols: o, river water blank; A, 120 mg/i vitamin-free casamino
acids as a control; •, 80 mg/l glucose; A, 80 mg/l glucose, KH2PO4 (0.33 mg/l
phosphorus) and K2HPO4 (0.33 mg/l phosphorus); Q, 80 mg/l glucose, (NH4)2 SO4
(3.33 mg/l nitrogen) and KNO3 (10 mg/l nitrogen); •, 80 mg/l glucose, K2HPO4,
KH2 PO4 and KNO3 (nitrogen and phosphorus in same amounts as above). The
glucose, K2HPO4f KH2PO4, (NH4)2SO4 and KNO3 were added to give the same
levels of carbon, phosphorus, and nitrogen as found in the casamino acids control.
-------
85
presence of glucose was observed with either polluted or unpolluted river water. This effect was
more pronounced with the unpolluted (Fig. 9) than with the polluted water (Fig. 8)., because it
altered both the extent of DO depletion and the time span, while only the time span was changed
in the polluted water. These results also suggested that a portion of the bacterial population was
not active in DO depletion with glucose as the substrate, possibly because the necessary growth
factors were not provided.
The effect of added nitrogen and phosphorus on DO depletion, with ethyl alcohol (Fig. 10) and
sodium acetate (Fig. 11} as the substrates, was studied in unpolluted water. The carbon content of
both substrates and the amount of nitrogen and phosphorus added were the same as in the
vitamin-free casamino acids control. Both of these substrates were even more poorly utilized for
DO depletion than was the glucose (Fig. 9) without the addition of nitrogen and phosphorus.
Again, as with glucose, the addition of nitrogen and phosphorus enhanced the utilization of ethyl
alcohol and sodium acetate. The DO depletion with the vitamin-free casamino acids control was
still greater even though the utilization in the presence of these substrates was enhanced. This is
added support for the role of growth factors in the metabolic activity of the bacterial population.
Both yeast and beef extracts contained slightly less ammonia nitrogen than did the vitamin-free
casamino acids. The addition of ammonia nitrogen had no effect on the utilization of DO with
either extract in polluted or unpolluted water, since the results were identical to those shown in
Figures 5 and 6.
40 60 80
TIME (HOURS)
FIGURE 10 Effect of ethyl alcohol on dissolved oxygen depletion in unpolluted sub-Arctic river
water when incubated at 10° C in the presence and absence of added inorganic
nitrogen and phosphorus. Symbols: o, river water blank; A, 120 mg/l vitamin-free
casamino acids as a control; •, 60 mg/l ethyl alcohol; D, 60 mg/l ethyl alcohol.
K2HPO4 (0.33 mg/l phosphorus), KH2PO4 (0.33 mg/l phosphorus), (NH4)2SO4
(3.33 mg/l nitrogen) and KNO3 (10 mg/l nitrogen). The ethyl alcohol, K2HPO4,
KH2PO4, (NH4)2SO4, and KNO3 were added to give the same levels of carbon,
phosphorus and nitrogen as found in the casamino acids control.
-------
86
20 tO 60 80
TIME (HOURS)
100 120
FIGURE 11 Effect of sodium acetate on dissolved oxygen depletion in unpolluted sub-Arctic
river water when incubated at 10° C in the presence and absence of added inorganic
nitrogen and phosphorus. Symbols: o, river water blank; A, 120 mg/l vitamin-free
casamino acids as a control; •, 180 mg/l sodium acetate; A, 180 mg/l sodium acetate,
K2HPO4 <0.33 mg/l phosphorus), KH2PO4 (0.33 mg/l phosphorus). (NH4)2SO4
(3.33 mg/l nitrogen) and KNO3 (10 mg/l nitrogen). The sodium acetate, K2HPO4,
KH2PO4, (NH4)2SO4 and KNO3 were added to give the same level of carbon,
phosphorus, and nitrogen as found in the casamino acids control.
Effect Of Sewage Treatment Plant Effluents On DO Depletion In Unpolluted Subarctic River
Water
The primary sewage treatment plant effluent contained 24 mg/l ammonia nitrogen, 0.01 mg/l
nitrite nitrogen, 0.15 mg/l nitrate nitrogen, 3.4 mg/l orthophosphate phosphorus and 235 mg/l
chemical oxygen demand (COD). This effluent was added to unpolluted river water in an amount
which gave a final COD of 59 mg/l. These results are shown in Figure 12. Oxidizable substrate,
growth factors and inorganic nutrients in the effluent permitted rapid DO depletion at all
incubation temperatures. This DO depletion was more rapid than with a high level of vitamin-free
casamino acids (Fig. 4). Since the indigenous population in the river water had no discernible
effect on DO depletion at any incubation temperature, it appeared that the effluent had a bacterial
population capable of rapid and extensive activity.
The effect of effluent from an activated sludge sewage treatment system on DO depletion in
unpolluted water is shown in Figure 13. Effluent from the activated sludge system operating at 0°
- 1.0° C was added to unpolluted river water, giving a final COD of 16 mg/l. The results showed
that DO depletion activity increased with increasing incubation temperature. This suggested that
either a change in growth factor requirements or different enzyme systems made more substrate
available for utilization at the higher incubation temperatures. The bacterial population in the river
water appeared to have some effect on the extent of DO depletion at 10° and 20° C, since the rate
and extent of depletion was increased when the effluent was incubated in river water.
-------
87
40 eo ao co
TIME (HOURS)
120 140 I6O
FIGURE 12 Effect of incubation temperature on dissolved oxygen depletion when effluent from
the Fairbanks, Alaska city primary sewage treatment plant was added to unpolluted
sub-Arctic river water. Symbols: o, 25% effluent and 75% river water; •, 25%
effluent and 75% sterile glass distilled water; •, 25% sterile glass distilled water and
75% river water.
ao ieo
160 200 240 280
TIME (HOURS)
320 360 4W480
FIGURE 13 Effect of incubation temperature on dissolved oxygen depletion when effluent from
a 0° • 0.5° C bench scale activated sludge sewage treatment system was added to
unpolluted sub-Arctic river water. Symbols: A, 25% effluent and 75% river water; o,
25% effluent and 75% sterile glass distilled water; •, 25% sterile glass distilled water
and 75% river water.
Effect Of Incubation Temperature On DO Depletion In Arctic Water In The Presence Of A
Complex Organic Substrate
Vitamin-free casamino acids at a concentration of 120 mg/l were used as the substrate for DO
depletion studies in arctic river water (Sag River). The results, given in Figure 14, showed a lag
phase at all incubation temperatures before DO depletion began. The lag phase was extremely long
at the lower temperatures, particularly at 0° C. However, the extent of DO depletion did not
appear to be temperature dependent. It has been shown previously (Gordon, unpubl) that the Sag
-------
88
80 120
160 200 240 280 500 540 580 620 660 700
TIME (HOURS)
FIGURE 14 Effect of incubation temperature on dissolved oxygen depletion when 120 mg/1
vitamin-free casamino acids were added to unpolluted Arctic river water. A river
water blank (o) was incubated atO°, 10° and 20° C.
River had a large population of heterotrophic bacteria capable of growth at low temperatures.
Since only one large volume sample was available from the Sag River, the reason for the extended
lag phase remains to be determined.
The same substrate concentration and incubation temperatures made it possible to relate the
results from both the Sag {Fig. 14) and Chena (Figs. 3 and 4) rivers. One outstanding point was the
relative time before the start of DO depletion. There was a lag phase only at the 0° C incubation
temperature with samples from the polluted location on the Chena River, and the lag phase was
apparent only at 0°, 5° and 10° C with samples from the unpolluted location. An extended fag
phase at all temperatures was observed with samples from the Sag River. A point of similarity with
all samples was that the extent of the DO depletion did not appear to be temperature dependent.
The results shown in Figures 3. 4 and 14 did not fit either the first or second order kinetic forms,
so a rate constant was not obtained. Approximate rates (mg/l/hr) of DO depletion were obtained
directly from the depletion curves, and the results are presented in Table 4. it must be stressed
that these results are approximations and have value only in the context of this study. It would
seem reasonable to have found the highest rates of DO depletion with polluted Chena River water.
However, unpolluted Chena River water apparently gave higher rates than the polluted equivalent
at 15° and 20° C. The rates from both Chena River samples were nearly the same at 0°, 5° and 10°
C. The sample from the Sag River gave lower rates at 10°, 15° and 20° C than either Chena River
sample. This suggested that the bacteria from the Sag River were more adversely affected by the
higher incubation temperatures than those from the Chena River. Additional support for this
suggestion was the nearly equal rates found at 15° and 20° C with Sag River water. The results
showed that the source of the sample had little or no effect on the rate of DO depletion at 0° and
5° C. This suggested that all or part of the bacterial population from each source had the same
ability to utilize an organic substrate at low temperatures.
-------
89
TABLE 4
Comparison of the Rate of Dissolved Oxygen Depletion When a Substrate
Was Added to Arctic and Subarctic River Water Samples
Incubation
Temperature Rate of Dissolved Oxygen Depletion (mg/l/hr)
Subarctic River Arctic River
Polluted Sample Unpolluted Sample Unpolluted Sample
Location" Location0 Location
20° 1.36 1.73 0.62
15° 0.92 1.13 0.65
10° 0.63 0.53 0.35
5° 0.22 0.26 0.20
0° 0.22 0.25 0.20
a. 120 mg/1 Vitamin-Free Casamino Acids (Difco) was added to each river water sample.
b. The polluted location was below a reach of the river receiving raw domestic sewage and
effluents from primary sewage treatment plants.
c. The unpolluted location was upstream from any source of domestic or industrial
pollution.
DISCUSSION AND CONCLUSIONS
A subarctic (Chena) river had a population of heterotrophic bacteria capable of growth at 0° C on
a complex medium. With dissolved oxygen (DO) depletion as the measurement, there appeared to
be little metabolic activity in a closed, stationary river water system. When vitamin-free casamino
acids were added to the stationary system, there was rapid and extensive DO depletion at all
incubation temperatures (Fig. 4). The rate of DO depletion appeared to still be increasing at the
lower incubation temperatures when the oxygen was exhausted, which suggested the maximum
rate had not been reached. Thus, oxygen may have been limiting.
Jezeski and Olsen (1961) found that shake cultures increased growth rate and maximum growth
level of Pseudomonas fluorescens at 4° and 10° C as compared to stationary cultures. In the shake
cultures, oxygen was no longer a limiting factor, and the bacterial cells were kept in a constantly
changing micro-environment which removed metabolic end products and brought the cells in
contact with new substrate. Such a dynamic system more nearly simulates environmental
conditions in a river than does a stationary system. The indigenous bacteria had the potential for
rapid metabolic activity in a stationary system. Next, a dynamic system must be studied to more
accurately assess the role of these bacteria in the natural environment and the effects of added
substrates, such as sewage effluents.
Metabolic activity observed with the protein derivative, vitamin-free casamino acids, as the
substrate was not as rapid as with yeast or beef extract (Fig. 6). These extracts contained growth
factors and carbohydrates in addition to proteinaceous material. It was apparent that one or more
growth factors were responsible for the increased metabolic activity. However, it is not known
whether the growth factors enhanced the activity of all or part of the bacterial population, or
-------
90
whether a portion of the population had an absolute growth factor requirement. An understanding
of the role of growth factors is necessary as an aid in developing design criteria for sewage
treatment plants that will provide sewage effluents that minimize the demand for DO.
In a review of psychrophilic bacteria, Ingraham and Stokes (1959) pointed out that they could
carry out nearly all metabolic activities at low temperatures, but at a slower rate than at higher
temperatures. Several psychrophilic and mesophilic Arthrobacter species were studied for effect of
temperature on growth by Roth and Wheaton (1962). Rather than a sharp cut-off point, there was
a continuous gradation with a decreasing lag phase at 0° C and an increasing one at 37° C. The
longest lag phase they measured at 0° C was about 300 hours before the start of fairly rapid
growth. They concluded that the number of generations of a specific bacterium was not
temperature dependent, but the time to attain a certain number was extended at lower
temperatures.
The decreasing rate of metabolic activity with decreasing incubation temperature which was
reported previously (Ingram and Stokes, 1959) appeared to be borne out by the results reported
here (Table 4). This was true with samples from above and below the polluted reach of the Chena
River. Several significant effects on metabolic activity in the samples were noted after the Chena
River had flowed through the polluted reach. The lag phase before the start of DO depletion was
much shorter (Fig. 3) than with samples from above (Fig. 4), which resulted in a much shorter
elapsed time from the start of incubation until all of the DO had been utilized. The apparent effect
of growth factors on the rate of DO depletion was reduced (Figs. 5 and 6), and glucose was more
effectively utilized as a substrate (Figs. 8 and 9). These effects on metabolic activity indicated that
raw sewage and primary treatment plant effluents added a high level of organic substrates, growth
factors, nitrogen and phosphorus to the river water. In addition to the nutrients, the results
presented in Figure 12 showed that bacteria capable of rapid metabolic activity at low
temperatures were present in the primary treatment plant effluent. Because of these factors, raw
sewage and primary effluents would probably significantly increase the DO demand under ice
cover.
McDonald, et al. (1963) found proteolytic bacteria in arctic littoral and marine sediments. They
found that proteolytic enzymes were highly active at low temperatures and proposed that these
enzymes might be significant in protein degradation in the Arctic. Rapid and extensive DO
depletion was found in Chena River water at low temperatures with protein derivatives as the
substrates. This suggested that proteolysis is one of the major metabolic activities of the bacteria in
the Chena River.
A large variety of proteolytic bacteria have been found in sewage treatment systems (Green,
1964). Since proteolytic activity has been found at low temperatures, there are probably similar
bacteria present in sewage treatment systems operating at low temperatures. Support for this
suggestion was obtained from an activated sludge sewage treatment system operating at 0° C. This
system reduced the DO requirements of domestic sewage to a level that appeared to be of minimal
influence in Chena River water at 0° C (Fig. 13).
Earlier work with pure cultures of Pseudomonas fluorescens showed that glucose was a poor
substrate for DO utilization at low temperatures (Jezeski and Olsen, 1961) and that the generation
time was much longer than with vitamin-free casamino acids as the substrate (Olsen and Jezeski,
1963). This may have resulted from a change in the metabolic pathway of gluscose utilization at
-------
91.
low temperatures (Jezeski and Olsen, 1961; Palumbo and Witter, 1969b), or that more glucose was
consumed for cell maintenance at low temperatures (Palumbo and Witter, 1969b). Inoue, et al.
(1967) found that acetate-oxidizing bacteria were vital in self-purification of rivers.
When substrates (glucose, sodium acetate and ethyl alcohol) that did not contain nitrogen,
phosphorus or growth factors were added to the closed, stationary river water system, a low level
of metabolic activity resulted. Adding nitrogen and phosphorus resulted in a marked increase in
activity (Figs. 9, 10 and 11). Even with these nutrients present, the rate of DO depletion was
greater with the vitamin-free casamino acids. This suggested that a portion of the bacterial
population either required additional growth factors, or was not capable of utilizing these
substrates. Even though bacteria capable of utilizing these substrates were present in the Chena
River, activity would probably be at a low level because of the limited amount of nitrogen and
phosphorus present under natural conditions.
Bacteria were found to have an important role in the cycling of phosphorus in the aquatic
environment (Philips, 1964), and both phosphorus and nitrogen appeared to be effective in
limiting metabolic activity in the Chena River under natural conditions (Fig. 9). Nitrogen and
phosphorus present in raw sewage and primary sewage treatment plant effluents reduced the
limiting effect of these nutrients (Fig. 8). Therefore, a method must be found to control nitrogen
and phosphorus in effluents entering arctic or subarctic waters. Barth, etal. (1968) demonstrated
that it is feasible to remove both nitrogen and phosphorus on a pilot plant scale using a combined
chemical-biological removal'system. Since several methods are available (Nesbitt, 1969), the state
of the art of phosphorus removal is probably much more advanced than nitrogen removal. Perhaps
the initial efforts should be directed toward adapting a phosphorus removal method.
Throughout this study, nitrogen was supplied in the form of ammonia and nitrate at the levels
present in the vitamin-free casamino acids. It is necessary to determine if the nitrogen form has
any effect, and what concentration is actually required. This should aid in determining what could
be done to control the effect of nitrogen on receiving waters.
Results obtained with arctic river water were far too limited to be conclusive. However, there are
some general similarities between the arctic (Fig. 14) and subarctic rivers (Figs. 3 and 4). More
detailed study is necessary before the effects of pollutants on arctic rivers can be defined.
It is becoming increasingly obvious that the 5 day, 20° C BOD (biochemical oxygen demand) has
very limited usefulness in the arctic or subarctic because receiving waters rarely reach this
temperature. Previous studies by Murphy and Miller (1968), Reid and Benson (1966), and Reid
(1968) showed that a 20 day BOD, incubated at a low temperature with receiving water or seed
culture acclimatized at a low temperature, gave more realistic results with raw sewage. The results
presented here (Figs. 12 and 13) showed that incubation temperature and diluent had an effect on
DO depletion with sewage treatment plant effluents. Therefore, it is suggested that the receiving
water should be used as the diluent and the incubation temperature should be at or near the
temperature of the receiving water.
-------
92
ACKNOWLEDGMENTS
Mrs. Becky L. Quimby for her able assistance in the laboratory.
Mr. Ernest W. Mueller and his staff for providing the chemical data presented here.
Mr. Michael A. Angelo for assistance in the mathematical treatment of the data.
Mr. Sidney E. Clark for help in developing the equipment used for aeration and temperature
equilibration of the samples.
Use of product and company names is for identification only and does not constitute endorsement
by the U.S. Department of the Interior or the Federal Water Quality Administration.
REFERENCES
Adams, J. C. and Stokes, J. L. (1968) Vitamin requirements of psychrophilic species of bacillus, J.
of Bacteriology, 95, p 239.
American Public Health Association (1965) Standard Methods for the Examination of Water and
Wastewater, 12th Edition, American Public Health Association, Inc., New York.
Association of Official Agricultural Chemists (1965) Official Methods of Analysis of the
Association of Agricultural Chemists, 10th Edition, Association of Official Agricultural
Chemists. Washington, D. C.
Barth, E. F., Brenner, R. C. and Lewis, R. F. (1968) Chemical-biological control of nitrogen and
phosphorus in wastewater effluent, J. Water Poll. Control Fed., 40, p 2040.
Boyd, W. L. (1958) Microbiological studies of arctic soils. Ecology, 39, p 332.
Boyd, W. L. and Boyd, J. W. (1967) Microbiological studies of aquatic habitats of the area of
Inuvik, Northwest Territories, Arctic, 20, p 27.
Drachev, S. M. (1964) The oxygen regime and the process of self purification in reservoirs with
retarded discharge. Advances in Water Pollution Research, 1, p 17, The MacMillan Company,
New York.
Farrell, J. and Rose, A. H. (1965) Low temperature microbiology. Advances in Applied
Microbiology, 7, p 335.
Farrell, J. and Rose, A. H. (1967) Temperature effects on microorganisms. Annual Review of
Microbiology, 21. p 101.
-------
93
Federal Water Quality Administration (1969) FWPCA Methods for Chemical Analysis of Water
and Wastes, FWQA, Division of Water Quality Research, Anal. Qual. Cont. Lab., Cincinnati,
Ohio.
Fournelle, H. J. (1967) Soil and water bacteria in the Alaskan subarctic tundra, Arctic, 20, p 104.
Frey, P. J. (1969) Ecological changes in the Chena River, FWQA Publication, Alaska Water
Laboratory, College, Alaska.
Unpubl data, FWQA, Alaska Water Laboratory, College, Alaska.
Gordon, R. C., Unpubl data, FWQA, Alaska Water Laboratory, College, Alaska.
Green, S. R. (1964) Proteolysis and proteolytic organisms, In Principles and Applications in
Aquatic Microbiology, p 430, John Wiley & Sons, Inc., New York.
Ingraham, J. L. and Maaloe, O. (1967) Cold sensitive mutants and the minimum temperature of
growth of bacteria. In Molecular Mechanisms of Temperature Adaptation, p. 297, American
Association for the Advancement of Science, Washington, D. C.
Ingraham, J. L. and Stokes, J. L. (1959) Psychrophilic bacteria. Bacteriological Reviews 23, p 97.
Inoue. Z., Honda, A. and Ishii, R. (1967) The role of acetic acid degrading bacteria in
self-purification of freshwater streams, J. of Fermentation Technology, 45, p 570.
Jezeski, J. J, and Olsen, R, H. (1961) The activity of enzymes at low temperatures, p 139,
Proceedings of the Low Temperature Microbiological Symposium, Campbell Soup Company,
Camden, New Jersey.
Lotspeich, F. B., Unpubl data, FWQA, Alaska Water Laboratory, College, Alaska.
McDonald, I. J., Quadling, C. and Chambers, A. K. (1963) Proteolytic activity of some cold
tolerant bacteria from arctic sediments, Canadian J. of Microbiology, 9, p 303.
Miller, A. P. (1967) The biochemical bases of psychrophily in microorganisms, a review, Institute
of Water Resources, University of Alaska, College, Alaska.
Mueller, E. W., Unpubl data, FWQA, Alaska Water Laboratory, College, Alaska.
Mulder, E. G. (1964) Arthrobacter, In Principles and Applications in Aquatic Microbiology, p 254,
John Wiley & Sons, Inc., New York.
Murphy, R. S., Asce, M. and Miller, A. P. (1968) Waste induced oxygen uptake of an Alaskan
estuary, J. Sanitary Engineering Div., A.S.C.E., 94, p 345.
Nesbitt, J. B. (1969) Phosphorus removal - the state of the art, J. Water Poll. Cont. Fed., 41, p
701.
-------
94
Olsen, R. H. and Jezeski, J. J. (1963) Some effects of carbon source, aeration, and temperature on
growth of a psychrophilic strain of Pseudomonas fluorescens, J. of Bacteriology, 86, p 429.
Palumbo, S. A. and Witter, E. D. (1969a) Influence of temperature on glucose utilization by
Pseudomonas fluorescens, Applied Microbiology, 18, p. 137.
(1969b) The influence of temperature on the pathways of glucose catabolism
in Pseudomonas fluorescens, Canadian J. Of Microbiology, 15, p 995.
Pereira, J. N. and Morgan, M. E. (1957) Nutrition and physiology of Pseudomonas fragi, J. of
Bacteriology, 74, p 710.
Philips, J. E. (1964) The ecological role of phosphorus in waters with special reference to
microorganisms. In Principles and Applications of Aquatic Microbiology, p 61, John Wiley &
Sons, Inc., New York.
Prince, H. N., Beck, E. S., Cleverdon, R.C. and Kulp, W. L. (1954) The flavobacteria, I. nutritional
requirements, J. of Bacteriology, 68, p 326.
Reid, L. C., Jr. (1968) Design and operation considerations for aerated lagoons in the arctic and
subarctic, Arctic Health Research Center Rept. No. 102.
Reid. L. C., Jr. and Benson, B. E. (1966) Observations on aerated sewage lagoons in arctic Alaska,
presented at the Eighteenth Annual Convention of the Western Canada Water and Sewage
Conference, Regina, Saskatchewan, Canada.
Roguski, E. (1967) Inventory and cataloging of the sport fish and sport fish waters in the interior
of Alaska, Alaska Department of Fish and Game, Federal Aid in Fish Restoration. Annual
Report of Progress, p 243.
Roth, N. G. and Wheaton, R. B. (1962) Continuity of psychrophilic and mesophilic growth
characteristics in the genus Arthrobaeter, J. of Bacteriology, 83, p 551.
Schallock, E. W., Unpubl data, FWQA, Alaska Water Laboratory, College, Alaska.
State of Alaska (1967) Water Quality Standards for Interstate Waters Within the State of Alaska
and a Plan for the Implementation and Enforcement of the Criteria, Department of Health
and Welfare, Juneau, Alaska.
Stefaniak, O. (1968) Occurrence and some properties of aerobic psychrophilic soil bacteria. Plant
and Soil, 29, p 193.
Stokes, J. L. and Redmond, M. L. (1966) Quantitative ecology of psychrophilic microorganisms,
Applied Microbiology, 14, p 74.
Straka, R. P. and Stokes, J. L. (1960) Psychrophilic bacteria from Antarctica, J. of Bacteriology,
80, p 622.
-------
95
Wuhrmann, K., Ruchti, J., and Eichenberger. E. (1966) Quantitative experiments on
self-purification with pure organic compounds, In Advances in Water Pollution Research, 1, p
229., Water Poll. Cont. Fed., Washington, D. C.
-------
PREDICTION OF DISSOLVED OXYGEN LEVELS IN THE
SOUTH SASKATCHEWAN RIVER IN WINTER
Robert C. Landine
INTRODUCTION
Water pollution is a problem of growing importance. The general public is becoming increasingly
concerned over the desecration of our natural resources through pollution. In Canada, the
Provincial and Federal Governments have acknowledged the need for preserving and upgrading our
water resources and have introduced more effective legislation for pollution control.
The discharge of municipal and industrial waste water effluents into a river may lead to excessive
demands on the dissolved oxygen (DO) resources of the river. To maintain the river in a
satisfactory state and in proper ecological balance it is necessary to prevent the DO concentration
from falling too low.
The DO concentration is regarded as one of the most important parameters for measuring or
assessing the degree of pollution in a river. Water pollution control agencies always include
minimum allowable DO concentrations in their list of pollution control parameters.
UNITED STATES OF AMERICA
FIGURE 1 Location of south Saskatchewan River in Canada
96
-------
97
The 00 concentration at any point along a river depends on the rates of oxygen supply and
demand. The respiration oxygen demand is counterbalanced by the supply of oxygen through
atmospheric reaeration and photosynthesis. The supply and demand of oxygen are related through
the complex phenomenon of oxygen balance which is determined by an interplay of many
physical, chemical and biological factors. Another important reason why the DO parameter is so
frequently used in pollution control work is that it is possible to use mathematical equations or
models to represent (approximately) the various forces operating in the oxygen balance. It is
possible to formulate equations for predicting the DO level at any point along a water course
under various conditions.
The opportunity for renewal through reaeration and photosynthesis is greatly, if not entirely,
reduced during the winter ice cover period which lasts five to six months in northern climates. A
blanket of ice and snow over the water shuts out the light, thereby denying photosynthesis. The
ice also prevents natural reaeration of the water.
From the above paragraph it follows that the oxidation of organic matter in ice covered rivers will
result in a progressive reduction in the DO level as the river flows downstream. Therefore, if the
oxidation rate is sufficiently high and the river is long (long time of passage) the DO concentration
will eventually fall to undesirable levels. Another point is that the flows in unregulated streams in
regions with prolonged frost periods tend to be minimal during winter. Therefore, the most critical
DO levels may be experienced in winter rather than in summer. Almost all cases reported in the
literature are for critical conditions occurring in summer.
With the distinct possibility of minimum DO levels occurring in winter coupled with the need for
better methods of predicting DO levels under ice conditions, an investigation was made into the
prediction of DO levels in a prairie stream under such conditions. The study was problem-oriented;
it was conducted from the viewpoint that the results should be useful to a water pollution control
agency.
The South Saskatchewan River was selected for a case study. It is a fairly large river, with an
average annual flow of approximately 10,000 cfs, being one of two major branches which combine
to form the Saskatchewan River. The Saskatchewan River System, which is 1,300 miles long, has
its headwaters in the Rocky Mountains near the Alberta-British Columbia border and flows
easterly through Alberta, Saskatchewan and into Lake Winnipeg, in Manitoba, as shown in Figure
1. From Lake Winnipeg, which is a large lake with an area of 9,000 mi2, the water flows into
Hudson Bay via the Nelson River. The Saskatchewan-Nelson River System is 1,600 miles long,
being the second longest in the country, and forms part of the Hudson Bay drainage area basin of
1,421,350 mi2, the largest in Canada.
The South Saskatchewan River is 900 miles long from the headwaters to the junction with the
North Saskatchewan River, approximately 20 miles east of Prince Albert, Saskatchewan. From the
profile shown in Figure 2 it is apparent that the river channel gradient is moderate across
Saskatchewan, being approximately 1.8 ft per mile.
The South Saskatchewan River was selected for a case study because 1) it is one of the most
important rivers in the region and one with which the writer had previous experience and 2) it was
very convenient, passing by the University of Saskatchewan in Saskatoon. As it turned out,
however, there were some aspects which rendered this river something less than an ideal choice.
-------
98
Soft
SosMchewon
Dtefenboker
"eOO—IKDO 1006 900 800 700 €00~
Distance above Lake Winnipeg in miles
FIGURE 2 Profile of Saskatchewan River system from headwaters to Lake Winnipeg
Diefenbaker Lake, the large reservoir created by Gardiner Dam, formed a logical upstream
boundary point, and the confluence with the North Saskatchewan River. 213 miles below the
Dam, was the downstream control point. This 213-mile reach of river received pollution loadings
from municipalities and industries and much greater loadings were anticipated in the future.
The 213-mile reach selected for a case study was subdivided into 11 reaches as shown in Figure 3.
DEVELOPMENT OF DO EQUATIONS (MODELS)
Ideally, a model should be developed from basic considerations, checked against field observations
and modified so that it will reproduce an existing situation. Then, presumably, the model could be
used to predict what would happen to the DO profile if certain changes, such as river flow and
waste loading, occurred in the river system.
The great amount of human and financial resources that would have been required precluded the
possibility of evaluating the various factors and coefficients from actual field surveys. Therefore, it
was decided that the coefficients would have to be determined from a literature search supported
by laboratory investigation and field surveys.
Time does not permit a description of the laboratory experiments conducted and the treatment of
data resulting therefrom. Only the pertinent results will be referred to as required in discussing the
oxygen balance situation for each reach.
Two worst winter situations were considered, viz., (1) future and (2) the conditions existing in
1968-69. Only the latter case will be considered in this paper. By worst situation is meant a
prolonged period of subzero weather coupled with low barometric pressure and calm conditions.
Two different river flows were considered, 4,500 cfs and 11,000 cfs. A flow of 4,500 cfs was
possible and would occur when one of the three turbines was operating at the hydroelectric station
at Gardiner Dam; this would be a minimum winter flow. A flow of 11,000 cfs occurs when all
-------
three turbines are in operation. Also, a flow of 11,000 cfs was being released from Gardiner Dam
at the time the writer made a field survey to check the DO level, and other parameters,
downstream from the Dam. It was therefore useful to include a flow of 11,000 cfs {although such
a high flow would not produce a worst condition) so that the predicted DO profile could be
checked against the observed profile. There were no tributaries and, thus, the same flow applied
over the entire reach.
The input waste loadings and station mileages are given in Table 1.
A discussion of the important factors considered in developing the equations to calculate the DO
profile is given below.
INITIAL CONDITIONS
The initial DO concentration of the water entering the river at the exit from the Dam was taken as
11.5 ppm, the value observed on two occasions.
The initial value of the equilibrium saturation value for oxygen dissolved in water at 0° C was
taken as 14.63 ppm based on the data presented by Montgomery, et al. (1964). Only one
temperature was used as it was assumed that the water temperature in both the open water reaches
and the ice covered reaches could be taken as 0° C. Only a minimal error was inherent in this
assumption. A station pressure correction was required. Based on observations made by the
NBNWN
PWNCE
ALBERT
Station
• For ry (name as show)
3Om MM below Gardimr Dam
R««ch rutter
l"«20mil»«
UNITED STATES OF AMERICA
COTEAU CREEKf
POWER STATIC
GARDINER DAM
: WEFENBAKER
FIGURE 3 South Saskatchewan River from Gardiner Dam to confluence with North
Saskatchewan River
-------
100
Saskatchewan Research Council (1969) a typical winter low pressure would be 698 mm of
mercury. No allowance was made for the differences in station pressure due to the difference in
elevation of the open water reaches at the Dam and 70 miles downstream at Saskatoon. The
difference in station pressures at the two locations was so small that it was satisfactory to consider
only one saturation value, i.e., 14.63 x 698/760 = 13.45 ppm for both locations.
TABLE 1
Station Description and Waste Loadings
Station Reach Station BODS Input in
Station Name Number Number Mileage Pounds Per Day
Gardiner Dam 1 0.0 0
1
End Open Water 2 5.0 0
Outlook 3 18.5 280
3
Q.E. Power Station 4 71.5 0
Weir Upstream 5 75.9 0
Weir Downstream 6 75.9 0
&
Present City Outfall 7 76.7 30,000
IPCO-Armour 8 80.9 300
8
Clarkboro Ferry 9 93.9 0
ft
Gabriel's Perry 10 124.0 0
10
St. Louis 11 155.0 40
Weldon Ferry 12 202.0 0
Reach 1, Gardiner Dam To End Open Water, Mile0.0 To 5.0
The river remains open, even during the coldest weather, below Gardiner Dam as a result of the
water being discharged from the hydroelectric station at the Dam.
The lowest rates of reaeration would occur during low pressure, calm conditions. It was assumed
that in the worst situation both of these conditions could occur in the open water reach 1 and
again when the corresponding volume of water arrived at the open water reaches in Saskatoon.
The minimum area of open water exposed to the atmosphere would occur at the end of a
prolonged period of subzero weather. Under these conditions it was estimated that the river would
be open over the entire channel width for a distance of 5 miles below the Dam.
The following equation developed by Owens, et al. (1964)
k2 = 9.41V°-67rf1-85 <1)
where k2 = reaeration coefficient day , base 10
-------
101
V = river velocity, fps
h = river mean depth, ft
was used to calculate the reaeration coefficient. No reduction was made for the effect of dissolved
salts, detergents or other contaminants since the concentration of all of these would be low,
resulting in a negligible correction to k2.
As is evident from Equation 1, k2 is a function of velocity and depth. Since the input data was in
terms of flow, Q, rather than h or V, it was necessary to express the latter two variables as a
function of Q. By plotting the data presented by Tywoniuk (1969) the following equations of best
fit were derived for reach 1
V = 0.11Q0'32 (2)
and
h = 0.072Q0'47 (3)
Likewise, the data given by Schriek (1963) was used to derive the following equations for reach 4,
which was partly ice free
V = 1.55Q0'77 (4)
h = 3.1 OQ°-095 (5)
It may be noted here that a relationship giving k2 as a function of Q may be obtained by
substituting Equations 2 and 3, or 4 and 5, into Equation 1. This will result in the following
equations for reach 1 and reach 4
k2 = 286Q'0'67 (reach 1) (6)
k2 = 0.015Q0'34 (reach 4) (7)
•J/o
From Equation 6 it is apparent that for reach 1 k2 is inversely proportional to Q , i.e., k2
increases as Q decreases which is very helpful from the water quality point of view. However, the
opposite situation is true for reach 4 which is atypical due to a weir located at the downstream end
of the reach. In this reach the reaeration rate increases as the flow increases.
Past experience (Lackie and Sparling, 1955; Sparling, 1957; Bouthillier, 1968; SWRC, unpubl;
Landine, 1962) with prairie rivers under ice cover conditions has shown that there is always a
certain background or residual BOD5 of the order of 1 to 2 ppm, in waters far removed from
significant waste water discharges. For example, the BODS of South Saskatchewan River water
upstream from Saskatoon was 1.0 ± 0.8 ppm (based on 18 samples collected in the months of
December to February from 1957 to 1969). The only significant sources of municipal or industrial
pollution were located hundreds of miles upstream.
The explanation of a residual BOD is believed to be related mainly to the slow oxidation of humus
and, to a lesser degree, to the oxidation of anaerobic end products. Humus refers to that portion
-------
102
of the organic matter that is very resistant to biological decomposition and could yield an
essentially constant daily BOD for several days if the amount of humus were sufficient and the
period of time involved not excessive. Periods of several months have been reported for complete
stabilization of polluted river water even at the normal temperature of 20° C (Gannon, 1963;
Garneson, 1958).
The humus material has its origin in 1) the waste water effluents discharged from industries and
municipalities in Alberta (mainly) and Saskatchewan and 2) the variety of organic materials
entering the river through land drainage. Land drainage occurs only during summer. However, due
to the long flow-through time in Diefenbaker Lake {1 year storage capacity) the effect is felt
throughout the winter period both in the lake and in the water released from it.
There would be a certain amount of anaerobic decomposition in the vicinity of the benthic
growths or muds; this would be particularly true in the lake. The oxidation of anaerobic end
products creates an oxygen demand.
The residual oxygen demand (ROD) was treated as a constant demand, in ppm per day, applied
over the entire 213 mile reach. The magnitude of ROD was obtained from the above mentioned
BODS @ 20° C, (i.e. 1.0 ± 0.8 ppm) as follows
BODS @ 20° C
ROD ppm/day = (8)
5 days x FACT
r»o
where FACT is the factor for converting from BODS @ 20 C to the BODS @ 0 C for polluted
river water. From experiments conducted by Landine (1970) it was found that the magnitude of
FACT was between 3.0 and 3.5 for polluted river water samples. From Equation 8 the mean value
of ROD was 0.062 ppm per day; the upper and lower values were 0.120 and 0.011 ppm per day
respectively.
There were no municipal or industrial discharges in reach 1.
There was no need to consider organic toad contributions due to local run-off along the entire 213
mile reach because there is no run-off during severe winter conditions.
Photosynthesis was assumed to be insignificant in the oxygen economy of the river in the open
water reaches as well as the. ice covered reaches. This assumption was believed to be reasonable in
view of the low temperature, the few hours of daylight, and the low intensity of sotar radiation in
winter. Table 2 illustrates that the solar radiation is much lower in winter than in summer in the
area.
TABLE 2
Solar Radiation at Saskatoon. 1955 to 1967
(Saskatchewan Research Council)
Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sept. Oct. Nov. Dec.
102* 199 326 414 508 544 568 466 326 197 103 74
Hie average annual total = 115 kilolanglies
* Units are langlies/day, equivalent to g cal/cm2 day.
-------
103
After evaluating the various factors of importance in reach 1, the DO equation was written. The
deficit at the end of reach 1 is a function of two forces, viz., 1) reaeration and 2) the constant
ROD rate. The differential equation may be written as
dDt
= -K2Dt+ROD (9)
dt T
and integrated to yield
Dt = DQe-K2t + ROD(1 -e'^VlCa (10)
where
Dt = DO deficit at time t, i.e., at end of reach,
DQ = DO deficit initially, i.e., at beginning of reach,
t = time of flow in reach in days, obtained from Equation 2 and 4. (For ice covered reaches
the flow time was increased by 30% to allow for the reduced velocity of flow.)
K2 = reaeration coefficient, day" , base e, and
ROD = constant residual oxygen demand rate @ 0° C, ppm/day.
It was unnecessary to write a partial differential equation because with no photosynthesis there
would be no diurnal fluctuation in the DO concentration; also, the waste loadings were treated as
constant throughout the day. A term for dispersion was also neglected; dispersion has a negligible
effect on the calculated DO profile in freshwater streams (Dobbins, 1964).
Reach 2, End Open Water To Outlook, Mile 5.0 to 18.5
There was no reaeration and no waste loading in this reach. The change in DO concentration was a
function only of ROD. Equation 11 was used to calculate the deficit at the end of reach 2
Dt = DQ + ROD(t) (11)
where each term is as defined previously; a complete list of symbols may be found in Appendix 1.
Reach 3, Outlook to Queen Elizabeth Power Station, Mile 18.5 to 71.5
Since organic matter will remain in suspension if the stream velocity is more than 1.0 fps (Imhoff,
et al., 1956) it was assumed that sludge deposits need not be considered in this river in which the
velocity exceeds 1 fps. Furthermore, most of the solids which would settle out are removed in
primary sedimentation and no waste water may be discharged without first receiving at least
primary sedimentation or its equivalent (SWRC, 1967).
A small amount of municipal waste is contributed by the town of Outlook.
-------
104
The forces causing the DO deficit to increase in this reach are the ROD and the first-order BOD
arising from the Outlook sewage discharge. Based on field and laboratory tests conducted by
Landine (1970) there was no need to allow for a nitrification oxygen demand because nitrification
did not occur at 0° C.
The oxygen demand due to the waste water discharge may be calculated from Equation 12 which
is the usual first-order BOD equation.
yt = L d-e-'V) (12)
where y = oxygen demand after t days, in ppm,
L = ultimate first-stage BOD, ppm,
K. = first-order BOD rate constant, day , base e.
The sequence of calculations was as follows:
(1) the waste discharge in Ib BODS @ 20° C per day (raw data, see Table 1) was converted to ppm
BOD5 @ 20° C in the river by dividing by 5.38Q;
(2) the river BODS @ 20° C was converted to a BODS @ 0° C by dividing by FACT for which
values of 3.0 and 3.5 were used;
(3) the river BOD5 @ 0° C was converted to an ultimate first-order BOD at 0° C by Equation 12;
(4) The ultimate BOD from step (3) was added to the ultimate BOD remaining at the end of the
previous reach (which was zero for reach 3 but not for subsequent reaches) to get the total
ultimate BOD @ 0° C at the beginning of the reach;
(5) the BOD in the reach was calculated by Equation 12,
(6) the BOD calculated in step (5) was subtracted from the total ultimate BOD at the beginning of
the reach to calculate the ultimate BOD remaining at the end of the reach.
To do step (3) above it was necessary to know the value of KI . The exact value of Kt in the river
was not known. To overcome this problem a range of Kt values, from 0.02 to 0.6 day"1, was used
in calculations. The lowest value 0.02, is very low and approximates a linear or constant oxygen
demand rate whereas 0.6 is believed to be as high as can be expected. Bouthillier (1968) reported
that the oxidation rate constant may be as high as 0.6 in the North Saskatchewan River below
Edmonton; Cameron (1967) reported the presence of slimes on the bottom in this reach which
could have accounted, in part at least, for the high rate of oxidation.
The following equation was used to caluclate the DO deficit at the end of reach 3
Dt = DQ + ROD(t) + YREACH (13)
where YREACH = first-order BOD exerted in the reach due to waste water contributions.
-------
105
Reach 4, Queen Elizabeth Power Station to Weir, Mile 71.5 to 75.9
The cooling water from two power stations maintains a strip of open water along one side of the
river in this reach. The width of the open water strip depends on such factors as the river flow,
duration of severe cold spell and distance below cooling water discharge point. It was estimated
that under the worst conditions assumed for this study 15% of the river channel would be ice free.
This reach was treated as two rivers flowing side by side separated by an imaginary wall to prevent
mixing back and forth. Since the reach was just 4.4 miles long with 85% of the channel width
covered with ice the above assumption was believed to be reasonable.
Reaeration, first-order BOD and ROD were the important factors considered in the open water
river. The differential equation was
dDt
= KiL + ROD -K2Dt (14)
dt T
which may be integrated to
if t k- i if * \f t RO° \f t
D = D e~N2t + NILO (e jt - e~N2T) + (1 -e~N2T) (15)
For the ice covered strip of river the forces considered were ROD and first-order BOD, and
Equation 13 was used to calculate the deficit at the end of the reach.
Reach 5. The Weir, Mile 75.9
The amount of aeration realized at this ogee-shape weir was calculated by means of Equation 16,
after Gameson, et al. (1958).
r = 1 + 0.11ab{a + 0.046T)ht (16)
where r = ratio of DO deficit above and below weir,
a = coefficient dependent on nature of water,
b = coefficient dependent on nature of weir, and
ht = fall height at weir in feet.
It was estimated that the value of a would be 1.25 because the river was 'clean' at this point.
Values of the coefficient b for an ogee-shape weir were not found in the literature; two alternate
values, 0.5 and 0.75, were estimated for the value of b in the computations.
The water fall height, ht> must be known to use Equation 16. It was found that a semiiog plot of
h versus Q linearized (Fig. 4) to yield the equation
-------
106
!
IU
8
6
4
?
— *-
Z
T-
/
h=K)x
.
o-aii
s*-^
I4xl0'
' j
4Q
-.
Reduc
totx*
effec
COVft
•— — ..
ed fall due .
*water /
t of ice -/
r dowr
exreurr
4 6 8 10 12 14 16
Discharge x IO"3 cubic feet per second
FIGURE 4 Fall height at Saskatoon Weir
ht = 10x 10
-0.134X 10"4Q
(17)
Equation 17 was based on data for open water conditions and a correction was necessary to allow
for the reduced fall height due to the backwater effect of the ice cover below the weir. Limited
field observations suggested a correction of minus 2 ft so that the equation used to calculate ht as
a function of Q was
'4
ht = 10x10-°-134x10'Q-2
Reach 6, Weir To Present Saskatoon Outfall, Mile 75.9 To 76.7
(18)
It was estimated that the river would be 100% ice covered at mile 76.7 and that 10% of the river
channel width would be open water in reach 6. Since the flow time in this reach was very small,
the differential equation could be left in the differential form, without introducing serious error,
to calculate Dr The forces considered were (1) ROD, (2) first-order BOD and (3) reaeration (in
the open water strip). The equation used was
dDt = KjLdt - 0.1K2D dt + ROD dt
(19)
where dD is the change in DO deficit in passing through the reach during a flow time equal to dt.
Reach 7,8, 9, 10 & 11, Mile 76.7 To 213
All of these reaches were treated in the same fashion because they were completely covered with
ice. Waste loadings were considered when necessary. (See Table 1 for amount and location.) The
two factors of significance were first-order BOD due to waste water effluents and ROD; Equation
13 was used to calculate the deficits at the end of each reach.
The prediction of the DO profile along the 213 mile reach was done with the aid of a digital
computer.
-------
107
DISCUSSION OF COMPUTER SOLUTIONS OF DO EQUATIONS
It was not possible to include all of the computed output as this involved several hundred lines;
therefore only a few selected iterations have been included to illustrate the main points of interest
as discussed below.
To check the effect of changing the variables Q, FACT, b. K, and ROD, a suitable range of values
was used for each one. This was written into the computer programs, using a nest of DO loops,
resulting in many iterative calculations for the various combinations. By scanning through the
output (which involved less than 2 min. computer time) the reader was quickly able to determine
which factors had the greatest significance in regard to the minimum DO concentration. This was a
low cost and rapid method of evaluating numerous alternatives.
The first point of interest centered around the question of whether or not the predicted profile
agreed with the observed profile.
The output showed that, when Q = 11,000 cfs, ROD = 0.011 ppm/day, FACT = 3.0 or 3.5, b = 0.5
and for any value of K| from 0.02 to 0.6 day"1, the calculated DO concentration at station 12*
was 12.0 ppm, which agreed very closely with the value of 11.9 ppm found on March 4, 1969. For
the same conditions, except with ROD changes from 0.011 to 0.062 and 0.12, the calculated DO
concentration at station 12 was 11.65 and 11.3 ppm respectively. The predicted values for the
three different values of ROD and the observed values are shown in Figure 5.
At first glance it appeared that the predicted profile was poorer for the higher values of ROD (i.e.
0.062 and 0.12), considering the furthest station downstream as the governing or most critical
point. However, a closer look showed that the predicted profile downstream from Saskatoon
approximately paralleled the observed profile for the case where ROD = 0.12 ppm per day,
suggesting that this upper value of ROD was the most appropriate one to use.
The discrepancies between the predicted and observed concentrations (which were 0.8 ppm or
less) would have been reduced if the field survey could have been made under the same conditions
as were assumed for the model. For example, due to the mild weather prevailing during the field
survey (March 4 and 5, 1969) the area of open water in reaches 1,4 and 6 was greater than for the
worst conditions assumed in the model. Also, instead of the assumed calm conditions, there was a
breeze ranging from 5.9 to 22.0 mph (mean daily), prevailing for the period February 22 to March
5, 1969 (SRC, 1969). The effect of the wind would have been to increase the reaeration rate, a
phenomenon which has been reported by others (Anon., 1964; Bohnke, 1966). Finally, the
barometric pressure was 15 to 28 mm of mercury greater over this same period than was assumed
in the model. All three of the above factors resulted in higher reaeration rates than for the assumed
conditions in the model and thus helped to explain why the observed concentrations were higher
than the predicted DO concentrations.
Based on the above mentioned results, it was felt that the oxygen balance forces and equations
employed were sufficiently representative to be useful in making predictions for this river.
*Station 12 was at Weldon, 202 miles below Gardiner Dam, the last sampling station before this
river flowed into the Saskatchewan River.
-------
108
A number of conclusions or deductions could be made from the computed output for winter
conditions.
(1) The magnitude of the value employed for ROD had a more pronounced effect on the results
than did the magnitude of K,. For example, for the minimum flow situation of 4,500 cfs,
changing ROD from 0.011 to 0.12 changed the DO concentration at station 12 from 11.93 to
10.98 pprn, i.e., 1 ppm, whereas changing K, from 0.02 to 0.6 changed the concentration from
11.93 to 12.02 ppm, i.e., 0.1 ppm.
(2) The magnitude of K! had little effect on YREACH and, consequently, on the DO profile. As
illustrated in column 15 of Table 3, when the other variables were held constant, changing Kj
from 0.02 to 0.60 day resulted in the sum of the first-order BOD changing from 0.53 to 0.43
ppm which is minor. From column 13 it can be seen that the value of the ultimate BOD at the
beginning of a reach is much higher for the case of K! = 0.02 than for K, = 0.60. However,
YREACH changed only slightly because a high value of ultimate BOD was used in conjunction
with a low value of K, to calculate YREACH (using Equation 12) and vice versa, which produced
a compensation effect. This suggested that it is not necessary to have an 'exact' value of KI for
calculating the first-order BOD in the river provided that the BOD5 @ 0° C of the waste water
loadings are known.
(3) The oxygen added in open water reaches and at weirs is highly significant in the oxygen
economy of ice covered rivers. This affords an opportunity for the oxygen concentration to be
increased considerably in a short time. The oxygen thus added becomes available for the slow
oxidation of organic matter in subsequent ice covered reaches.
I
i
FIGURES
--
2
-
/
r-Stopeporc
\ ROD=(
=
X
0
— &
O
ate) 10
)062
1-
o-
==£
Open water
:brtiy open <
Complete ice
Observed D
Predated D
Predicted D
Predicted D
— . *~
=£=£=
Aeration at
Saskatoon
Weir
1
r-C, =13.45 ppm (mode*
— -_.
~~-^*
rObserved
^
•ater
cover
0 concentrator Mar* 4-5, 1969
0 cone entroti on ROD = 0.01 1 ppm/Jay
0 concentration ROD = OO62 ppm/Wo
0 concentration ROO=O.I2Opprn/da)
-*
*-°— ^.
-lee-cover
assumed 1
y
— •<
-,
>ottern
x mode)
0 25 50 75 100 125 150 175 2CX
River miles below Gardiner Dam
Comparison of observed and calculated DO levels for winter conditions, Q = 11,000
cfs.
-------
TABLE 3
Partial Computed Output for 1968-69 Winter Conditions
1
STA
DK =
0.0
5.0
18.5
71.5
75,9
75.9
76.7
80.9
93.9
124.0
155.0
202.0
DK =
0.0
5.0
18.5
71.5
75.9
75.9
76.7
80.9
93.9
124.0
155.0
202.0
2
I
0.02
1
2
3
4
5
6
7
8
9
10
11
12
0.60
1
2
3
4
5
6
7
8
9
10
11
12
3
BOD
0.
0.
280.
0.
0.
0.
30000.
300.
0.
0.
0.
0.
0.
0.
280.
0.
0.
0.
30000.
300.
0.
0.
0.
0.
4
DK
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.60
0.60
0.60
0.60
0.60
0.60
0.60
0.60
0.60
0.60
0.60
5
ROD
ROD
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
ROD
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
6
BWEIR
= 0.011
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
= 0.011
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
7
FACT
3.0
3.0
3,0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
8
C
11.50
12.09
12.09
12.06
12.08
12.51
12.52
12.50
12.43
12.29
12.15
11.93
11.50
12.09
12.09
12.06
12.08
12.51
12.52
12.46
12.33
12.16
12.08
12.02
9
D
FACT
1.95
1.36
1.36
1.39
1.37
0.94
0.93
0.95
1.02
1.16
1.30
1.52
FACT
1.95
1.36
1.36
1.39
1.37
0.94
0.93
0.99
1.12
1.29
1.37
1.43
10
V
= 3
1.10
1.10
1.10
0.68
1.10
1.10
1.10
1.10
1.10
1.10
1.10
= 3
1.10
1.10
1.10
0.68
1.10
1.10
1.10
1.10
1.10
1.10
1.10
11
H
.0
3.91
3.91
3.91
6.89
3.91
3.91
3.91
3.91
3.91
3.91
3.91
.0
3.91
3.91
3.91
6.89
3.91
3.91
3.91
3.91
3.91
3.91
3.91
12
F
1.48
1.48
1.48
0.38
1.48
1.48
1.48
1.48
1.48
1.48
1.48
1.48
1.48
1.48
0.38
1.48
1.48
1,48
1.48
1,48
1.48
1.48
13
14
ULTROB ULTROE
BWEIR
0.00
0.00
0.04
0.04
0.04
0.04
4.38
4.40
4.35
4.22
4.09
3.91
BWEIR
0.00
0.00
0.00
0.00
0.00
0.00
0.44
0.39
0.26
0.11
0.04
0.01
= 0.50
0.00
0.00
0.04
0.04
0.04
0.04
4.36
4.35
4.22
4.09
3.91
= 0.50
0.00
0.00
0.00
0.00
0.00
0.00
0.38
0.26
0.11
0.04
0.01
15
YREACH
0.00
0.00
0.00
0.00
0.00
0.00
0.02
0.06
0.13
0.13
0.19
0.53
0.00
0.00
0.00
0.00
0.00
0.00
0.05
0.12
0.16
0.07
0.03
0.43
16
TIME
C
0.25
0.67
2.62
0.35
0.00
0.04
0.21
0.64
1.49
1.53
2.32
Q
0.25
0.67
2.62
0.35
0.00
0.04
0.21
0.64
1.49
1.53
2.32
17
TO TIME
= 4500
0.00
0.25
0.91
3.53
3.88
3.88
3.92
4.12
4.77
6.25
7.78
10.10
= 4500
0.00
0.25
0.91
3.53
3.88
3.88
3.92
4.12
4.77
6.25
7.78
10.10
NOTE: DK and BWEIR correspond to Kj and b respectively in script.
o
CO
-------
110
The figures in column 8 of Table 3 show that the oxygen level increased by (12.09-11.50) = 0.59
ppm in just 5 miles of open water in reach 1, which is more than the cumulative first-order BOD in
all reaches. In the same column, the results show that the amount of oxygen added through
aeration at the weir was approximately equal to the cumulative first-order BOD of the wastes
added in all reaches. It may be noted that more DO would have been added in reach 1 and at the
weir if the deficit had been greater,
By consideration of the basic forces of oxygen supply and demand it was possible to write a series
of DO equations which satisfactorily represented the oxygen balance in the South Saskatchewan
River under winter conditions.
It was found that the magnitude of the first-order deoxygenation rate constant was not critical in
evaluating DO levels when the BOD5 of the waste water loadings @ 0° C was known. The results
illustrated that any opportunity for reaeration, as in open water reaches and at weirs, is very
beneficial in the oxygen economy of streams under winter, ice cover conditions.
ACKNOWLEDGMENTS
The author wishes to acknowledge that in writing this paper he has used a portion of the work
reported in his thesis which was supervised by Prof. T. H. Lackie at the University of
Saskatchewan, Saskatoon, Canada, and supported financially by the National Research Council
through Operating Grant A-3811.
REFERENCES
Anon (1964) Effect of Polluting Discharges on the Thames Estuary, Technical Paper No. 11, Her
Majesty's Stationery Office, London.
Bohnke, B. (1966) New method of calculation for ascertaining the oxygen conditions in
waterways and the influence of the forces of natural purification. Adv. in Water Poll. Res.,
Vol. l.p 157.
Bouthillier, P. H. (1968) Biological oxidation in ice covered rivers, Proc. Third Canadian Symp.
Water Poll. Res., Vol. 3, p 12.
Cameron, R. D. (1967) Bio-oxidation rates under ice cover in the North Saskatchewan River, M.S.
Thesis, Univ. Alberta, Edmonton, Alberta.
Dobbins, W. E. (1964) BOD and oxygen relationships in streams, J. Sanitary Engineering Div.,
ASCE, Vol. 90, No. SA3. p 53.
Gameson, A. L. H., Vandyke, K. G. and Ogden, C. G. (1958) The effect of temperature on
aeration at weirs. Water and Water Engineering, Vol. 62.
Gameson, A. L. H. and Wheatland, A. B. (1958) The ultimate oxygen demand and course of
oxidation of sewage effluents, J. Proc. Inst. of Sewage Purification, Part 1, p 106.
-------
111
Gannon, J. J. (1963) River BOD abnormalities. Department of Environmental Health, Univ.
Michigan, Ann Arbor.
Imhoff, K., Muller, W. J. and Thistlethwayte, D. K. B. (1956) Disposal of Sewage and Other
Water-Borne Wastes, Butterworths Sci. Publ., London.
Lackie, T. H. and Sparling, A. B. (1955) Sanitary survey of the Red River, Manitoba Dept. Health
and Public Welfare, Winnipeg, Manitoba.
Landine, R. C. (1962) Pollution in the South Saskatchewan River at Saskatoon, M.S. Thesis, Univ.
Saskatchewan, Saskatoon, Saskatchewan.
Landine, R. C. (1970) Prediction of dissolved oxygen levels in the South Saskatchewan River,
Ph.D. Thesis, Univ. Saskatchewan, Saskatoon, Saskatchewan.
Montgomery, H. A. C., Thorn, N. S. and Cockburn, A., Determination of dissolved oxygen by the
Winkler Method and the solubility of oxygen in pure water and sea water, J. Appl. Chem.,
Vol. 14. p 280.
Owens, M., Edwards, R. W. and Gibbs, J. W. (1964) Some reaeration studies in streams, Inter. J.
Air Water Poll., Vol. 8, p 469.
Saskatchewan Research Council (1969) Physics Div. Saskatoon, Saskatchewan.
Saskatchewan Water Resources Commission, Poll. Control Br. Regina, Saskatchewan.
Saskatchewan Water Resources Commission (1967) Regina, Saskatchewan, Water quality
management policy in the Province of Saskatchewan.
Schriek, W. (1963) River sedimentation at Saskatoon, M.S. Thesis, Univ. Saskatchewan,
Saskatoon, Saskatchewan.
Sparling, A. B. (1957) Sanitary survey of the Assinibine River, Manitoba Dept. Health and Public
Welfare, Winnipeg, Manitoba.
Tywoniuk, N. (1969) Unsteady flow simulation of the South Saskatchewan River, M.S. Thesis,
Univ. Saskatchewan, Saskatoon.
-------
112
APPENDIX I - NOTATION
a
b
BOD
BODS
cfs
DO
DO
fps
FACT
h
2
K2
L
mi
mm
ppm
Q
r
ROD
t
V
Y^EACH
-1
coefficient in weir equation
coefficient in weir equation
biochemical oxygen demand
5-day biochemical oxygen demand
cubic feet per second
dissolved oxygen
DO deficit at beginning of reach, ppm
DO deficit at end of reach, ppm
feet per second
factor for converting BODS @ 20° C to BOD5 @ 0° C
mean depth of river, feet
fall height at weir, feet
first-order deoxygenation rate constant, to base e day
first-order reaeration coefficient to base 10, day
first-order reaeration coefficient to base e, day
ultimate first stage BOD. ppm
mile
millimeter
parts per million
flow in cfs
deficit above weir/deficit below weir
residual oxygen demand, ppm per day
time, days
velocity in feet per second
BOD realized in time t
first-order oxygen demand exerted in reach due to waste water conditions
-------
POLLUTION
A BIOLOGICAL STUDY OF SOME RECEIVING
WATERS IN HOKKAIDO
Matsunae Tsuda, Toshiharu Watanabe
and Kozo Tani
INTRODUCTION
The (shikari River is about 365 km long and is the largest river in Hokkaido (Fig. 1). Its middle
and lower reaches receive wastes from manufacturing industries, a pulp mill, and coal mines.
Discoloration of the river water results from the pulp mill and mine waste while the river bed is
covered by the coal mine silt.
IOOO
« tbetsu
8 I96I f
1C
FIGURE 1 Average water flow of the (shikari River at four stations
SOURCES OF POLLUTION
Upper Reach of the I shikari
Near the city of Asahikawa (pop. 245,000) three tributaries - the Ushukubetsu River, the
Chubetsu River and the Biei River - join the Ishikari River. The waste of the Kokusaku pulp mill is
discharged into the Ushukubetsu River and that of the Godo Alcohol Company factory into the
Chubetsu River. The effluent of the Kokusaku pulp mill has a BOD of 56.9 ppm and that of the
Godo Alcohol a BOD of 3,998 ppm in January 1959 (Kubo and Kosaka, 1960).
113
-------
114
TABLE 1
Water Temperature, pH, Transparency of the Stations
Studied in the Ishikari River
A Sounkyo
B Motoshirakawa
C Furukawa
D Kinseibashi
E Shinbashi
F Kyokuseibashi
G Ushubetsu R., near Kokusaku
Pulp Mill Effluent
H Ushubetsu R., Midoribashi
I Ino
J Fukagawa
K Sorachi R., Higashi-takegawa
L Naie
M Shimokawa
N Ebetsu
O Ishikari-ohashi
P 500 m downstream from
paper mill effluent
Q Toyohira R., Ganraibashi
R Ishikari-cho
Water Temperature
<°C)
0.5
2.8
2.3
2.8
2.5
6.5
11.5
11.5
•0.5
0
1.0
1.0
1.0
2.8
7.0
0.5
1.0
PH
7.0
6.8
6.8
6.6
6.9
6.8
6.0
6.8
6.8
6.8
6.8
6.7
6.8
7.0
6.9
7.0
6.8
Transparency
(cm)
46
12
45
18
25
11
11
5
25
6
30
7
Middle Reach of the Ishikari
The Sorachi and the Yubari rivers receive coal mine effluents. The load of the suspended
substances from the mines is estimated to be about 940 tons per day. Mine drainage and coal wash
waste contain very fine coal dust which changes the water of the receiving river to a brown color
and covers its bottom with silt. The influence of the coal mine wastes to the river organisms is
physical - the deposited silt makes it difficult for organisms to survive. It ruins fish spawning beds
and destroys flora and fauna. This form of pollution must be considered apart from the ordinary
saprobic system classification.
Still another pollution source on the middle reach is an artificial fertilizer factory of the Toyo
Koatsu Company located at Sunagawa. From this there discharges effluent containing much
inorganic substances especially SO4r the waste being heavily toxic to fish.
-------
115
Organic Wastes Flowing in at the Lower Reach
A factory of the Kita-Nihon Paper Mill Company at Ebetsu discharges waste water of 45,000 tons
per day, its BOD being 176 ppm (Tamura, Fuji, Yuki, 1961). Sewage treatment effluent and other
sewage of the city of Sapporo (pop. 820,000), as well as industrial waste are discharged into a
tributary, theToyohira River.
BIOLOGICAL WATER QUALITY ASSESSMENT BY THE BOTTOM MICROORGANISMS
At 18 stations (from station A to station R) on the river, the microorganisms attached to stones,
support piling and other suitable substrate, were collected and studied at the laboratory.
Table 2 shows the organisms found at stations A through R.
Upper Reach
The upper reach of the Ishikari was a clear mountain stream, with oligosaprobic waters, that were
rich in algae and benthic fauna.
Station A (Sounkyo): A famous resort known for its hot springs.
Station B (Motoshirakawa): A small tributary.
Station C (Furukawa): 13 km downstream of Sounkyo.
At stations A, B and C, transparency was more than 45 cm. Many kinds of algae were found.
Effluents from the hot-bath resort at Sounkyo have no important effect upon the river water. The
benthic flora consisted of blue-green algae, green algae and diatoms. A chrysomonad, Hydrurus
foetidus, occurred at Station A and Bactrachospernum moniliforme at B; these species are both
cold-water kathorobic species. Among the diatoms, Ceratoneis arcus, Fragilaria capucina var.
lanceolata, Gomphonema oliuaceum were dominant. Stations A, B and C were typical
oligosaprobic waters.
Asahikawa City and Surrounding Area
Station E (Shinbashi): This station was located on the left bank of the Ishikari River several
hundred meters downward from the point where the Ushukubetsu River joins the main stream.
This tributary receives waste of the Kokusaku Pulp Mill. Station E was heavily polluted and had a
strong unpleasant odor. Here Chromatium colonies cover the stones. There was Fusarium growth
which occurs frequently in waters with strong organic pollution and the presence of oxygen.
Scheuring and Zehender (1962) write that Fusarium occurs more often in cellulose-containing
wastes than the sewage bacteria Sphaerotilus. Practically no algae was found at this station. This
station belonged to polysaprobic waters.
Station F (Kyokuseibashi): This station was on the right bank of the main stream, some hundreds
of meters downstream from the junction of the tributary Ushukubetsu River, and 1 km
downstream from the pulp mill, on the opposite side of the stream from the waste effluent, thus
-------
116
TABLE 2
Benthic Microorganisms Found at the Stations in the I shikari River
SPECIES STATION
ABCDEFGHI J KLMNOPQR
Bacteria
Zoogloea ramigera + $ +
SphaerotUus natans ******** + t + +
Chromatium sp. t
Fungi
Alternaria sp. * +
Fusariam sp. * *
Blue-green algae
Chroococcus minutus + t
Chlorogloea microcystoides *
Dermocarpa flahaultii * +
Xenococcussp. + +
Oscillatoria formosa ** + + +
Oscillatoria sp. +
Phormidium uncinatum $ t
Phormidium spp. t +
Lyngbya sp. +
Homoeothrix janthina * ** +
Diatom
Melosira varians + + + ++-f+4= + + + +
Cyclotella comta + +
Tabellaria fenestrata +
Diatoma hiemale + +
Ceratoneis areas #*»#^+ + + +
Fragilaria capucina t $ * +
var. lanceolate
Asterionella formosa +
Synedraulna $ + + ++ + ++ +
S. ulna var. Ramesi + * + + + +
S. capitellata f. striis + + + +
Achnanthes linearis + +
A. Biasolettiana %
Cyrosigma acuminatum +
Navicula mutica $
N. cryptocephala * + + +
N. exigua +
Cymbella ventricosa + + +++
C. turgidula var. nipponica + + +
C. tumida +
Gomphonema parvulum +
G. olivaceum $ + **
Hantzschia amphioxys
Nitzschia linearis
N.palea 4=
-------
117
TABLE 2 (confd)
SPECIES STATION
ABCDEFGHI J KLMNOPQR
(keen algae
Ulothrix subtilissima + * T
U. zonata * * + + + +
Hormidium Klebsii + + + *
Geminella crenulatocollis +
Stigeoclonium lubricum * +
Rhizoclonium Hookeri *
Blubochaete sp. £
Oedogonium sp. *
Closterium sp. +
Rhodophyceae
Batrachospermum moniliforme *
Chrysophyceae
Hydrants foetidus *
Ciliata
Vorticella campanula +
Carchesium polypinum * *
Carchesium sp. +
** very rich, * rich, * common, + rare, + very rare
unaffected. Many kinds of blue-green algae, green algae and diatoms were found. This station was
oligosaprobic.
Station G (Ushubetsu River, Kokusaku Pulp): This was near the out-fall of the waste from the
Kokusaku Pulp Mill on the right bank of the Ushukubetsu River. There was an unpleasant odor of
factory waste which included the smell of hydrogen sulphide. The river water was dark brown in
color. Stones were black-colored on their bottom surfaces due to sulphides. There were no aquatic
insects. Growth on the stones consisted of Zooglea andAItemariasp. (fungus), both of which are
polysaprobic species which live in slow-running streams. This station belonged to polysaprobic
waters.
Station H (Ushubetsu River, Midoribashi): This was also a station of the Ushukubetsu River, 1 km
downstream from the paper mill. It was situated on the opposite side of the factory, but evidently
influenced by the factory waste, as indicated by blackening of the bottom surfaces of the stones.
Attached algae consist of blue-green algae (Phormidium uncinatum) and diatoms.
Station I (tno) and Station J (Fukagawa): Both stations were on the left bank of the river.
Influence of the paper mill was stronger here than on the other bank. At both stations there was
an odor of pulp waste, brown water with reduced transparency, and the bottom surfaces of stones
blackened. Sphaerotilus and Fusarium were found concurrently with diatoms. Both stations were
a-mesosaprobic* (am+ means the class belonging to am, but more strongly polluted than the
average am).
-------
118
The Sorachi River and Sunagawa
In this reach the Ishikari River received the effluents of many coal mines. The wastes contained
almost no organic compounds. Their effect on the river organisms was mostly the physical effect
of siltation.
Station K (Higashi-takigawa): This station is at the Sorachi River, a tributary. Due to the coal mine
waste, the water was brown-colored, with low transparency. The river bed was covered by silt and
coal dust. Organisms were very scarce. Study of algae attached to stones covered by silt revealed
that the flora consisted only of diatoms. (Table 2).
Station L (Naie): Located on the left bank of the main stream, this was 20 km downstream from
the junction of the Sorachi River and 5 km downstream from the out-fall of the waste of the
Toyo-Koatsu factory. Transparency was reduced and the water was brown in color.
Lower Reach and the Toyohira River
Station M (Shimokawa): This station was 40 km downstream from Station L. At both Station L
and M, a few blue-green algae and diatoms were found. Sphaerotilus and fungus were also present,
but in small quantity. Both stations are a-mesosaprobic.
Station N (Ebetsu): This station was 1 km downstream from the junction of the Yubari River,
which carries a large amount of silt from coal mines into the Ishikari River. The river bed was
covered by silt. Only two species of diatoms were represented here, and no benthic animals were
found.
Station O (Ishikari-ohashi): This station was 1.5 km downstream from the junction of the Ebetsu
River. There was much silt on the river bed, however transparency was recovered (Table 1).
Micro-flora on stones consist of Zooglea. Sphaerotilus, blue-green algae and diatoms. The station
was cc-mesosaprobic+.
Station P: About 500 m downstream from the waste of the Kita-Nihon Paper Mill factory and also
the out-fall of cooling water of a heat-engine plant; its water was brown-colored, with an
unpleasant odor. Sphaerotilus, blue-green algae and diatoms were sparse. The station was
polysaprobic.
Station Q (Ganraibashi): This station was on the Toyohira River, a tributary which flows through
Sapporo City, the largest city in Hokkaido. The station was about 300 m downstream from the
night soil treatment plant. Algae on stones consist mainly of blue-green algae, Hormidiun sp.
(green algae), Zooglea and Sphaerotilus occurred too.
Station R (Ishikari-cho): This station was at Ishikari-cho, a town 5 km downstream from the
mouth of the river. Water was brown in color, with transparency of 7 cm. Ciliates and a few
diatoms were found. The diatoms consisted of such species as Synedra ulna, Nitzschia palea,
Melosira variens, and Navicula cryptocephala. The station was a-mesosaprobic*.
Figure 2 shows the water quality map made on the basis of the biological assessment by the
above-mentioned data on microorganisms.
-------
119
50 —
100 —
150-
200 —
230 -
FIGURE 2 Water quality map of the Ishikari River by biological assessment
BENTHIC ANIMALS
Tables 3 and 4 show the benthic animals found at Stations A through R.
Upper Reach
The upper reach of the Ishikari was rich in benthic fauna.
Station A (Sounkyo): Large numbers of benthic species were found; the water was clear. The
net-spinners coefficient (this equals the ratio of the biomass by weight of net-spinning caddis fly
larva to the biomass of total benthic animals) was 82%, showing that the community was stable,
i.e., at climax or near climax. Station A was oligosaprobic.
Station B (Motoshirakawa): Similar to Station A, oligosaprobic.
-------
120
TABLE 3
Benthic Animals Found in Three Stations in the Upper Reach
of the (shikari River
SPECIES STATION
Sounkyo Motoshirakawa Furukawa
ABC
Ephemeroptera
Paraleptophelebia sp. + +
Ephemerella basalis
Ephemerella ru fa
Ephemerella nigra + +
Ephemerella sp. nay £
Ephemerella sp. EC ± $
Ephemerella sp. ER ±
Baetis sp. +
Baetiella sp. +
Isonychia japonica +
Epeorus uenoi +
Epeorus latifolium +
Rhithrogena sp. +
Cinygma sp. + $
Plecoptera
Nemoura sp. * +
Amphinemoura sp. * t
Protonemura sp. £ t
Capnia sp. +
Megarcys ochracea +
Isoperla towadensis $ $
Hemiptera
Notonecta trigutta +
Sigara nigroventralis +
Neuroptera
Stalls sp. +
Trichoptera
Rhyacophila articulata +
Rhyacophila brevicephala +
Rhyacophila sp. RG £ £
Mystrophora sp. (Larva) + *
Mystrophora sp. (Pupa) * i
Stenopsyche griseipennis * •»• i
Arctopsyche sp. E + +
Hydropsyche ulmeri * ^
Hydropsyche sp. HB +
Molanna falcata + +
Neophylax ussuriensis t
Glyphotaelius admorsus +
Stenophylax sp. +
Goera japonica t +
Brachycentrus sp. ^
Micrasema sp. 4-
Coleoptera
Philorus ezoensis +
Tipula sp. + +
Antocha sp. i ^
Spaniotoma sp. * $
Atherix sp. + +
Turbellaria
Dendrocoelopsis sp.
-------
121
SPECIES
TABLE 4
Benthic Animals of Stations E - R
STATION
Ephemeroptera
Ephemerella sp. EC
konichia japonica Ulmey
Plecoptera
Isoperla towadesis Okamolo
Trichoptera
Neoseverinia crassicornis Ulmey
Coleoptera
ttybius apicalis Sharp, adult
Dipt era
Tipula sp.
Antocha sp.
Spaniotoma sp.
Tendipes sp.
Crustaceae
Gammarus nipponensis Veno
Asellus ripponensis Nichols
Oligochaeta
Tubifex sp.
Hirudinea
Mimobdella sp.
H
I
rich,
K
M O P Q R
common, + rare
Station C (Furukawa): Similar to Station A, oligosaprobic.
Asahikawa City and its Neighborhood
Station D (Kinseibashi): The water was clear; transparency was 46 cm. There were very few
animals due to the devastation of the bottom by gravel-dredging. Kubo et al. (1961), at this
location recorded larvae of some species of mayflies, stoneflies and caddisflies.
Station E (Shinbashi): This reach was extremely polluted by the waste of the Kokusaku Pulp Mill.
Mimobdella sp. was the only animal found here. Station E was polysaprobic.
Station F (Kyokuseibashi): Some hundreds of meters downstream from the junction of the
Ushukubetsu River, it was on the other side of the pulp mill, so there was little influence from the
waste. Ephemerella sp. EC and Isoperla towadensis, which are clear-water species, were collected
here. This station was oligosaprobic.
Station G (Ushubetsu River): This was located at the out-fall of the waste of the Kokusaku Pulp
Mill. No bent hie animals.
-------
122
Station H (Ushubetsu River, Midoribashi): Station H was located 1 km downstream from the mill.
Tendipes sp., Asellus sp., and Tubifex sp. were found, all of which are pollution-tolerant species.
This station was0-mesosaprobic~.
Station I (Ino): Tendipes, Tipula, Asellus, Mimobdella, and Tubifex were found. The station was
a-mesosaprobic*.
Station J (Fukagawa): Tendipes, Asellus, and Mimobdella were found. Besides clean-water insect
species such as Ephemerella sp. EC, Neoseverinia crassicornis were found.
Lower Reach
Due to inflow of silt from the Sorachi River, a tributary, the river bed was covered by silt and coal
dust. Almost no stones were found.
Station K (Higashi-takigawa): Station K was represented by Tendipes and Tubifex.
Station L (Naie>: At Station L Isonychia japonica, Gammarus, Asellus, Mimobdella and Tubifex
were found. Station L was /3-mesosaprobic".
Station M (Shimokawa): The bottom at Station M was covered by silt and Ilybius apicalis was
present.
Station N (Ebetsu): Silt and coal dust from coal mines characterized Station N; no benthic animals
were found.
Station 0 (Ishikari-Ohashi): Gammarus sp. was found at Station O.
Station P: Station P was located 500 m downstream from the paper mill. The water was brown,
and Tubifex and Entosphenus japonicus sp. were found. Lamprey were widely distributed in the
middle and lower reaches of this river. They were found even in polluted waters, and seemed to be
very tolerant of the pollution.
Station Q (Ganraibashi): Located on the Toyohira River downstream from Sapporo. The
transparency of the water was 3 cm. Tendipes, Asellus, Tubifex and Entosphenus japonicus were
found; Tubifex was very abundant. Station Q was a-mesosaprobic.
Station R (Ishikari-cho): The water transparency was 7 cm. The water color was brown and the
bottom was covered with mud. Tubifex was present. Station R was a-mesosaprobic*.
DISCUSSION
When we assess water quality biologically, we must consider two kinds of effects separately:
(a) The effect of organic pollution from municipal sewage and paper or pulp mills.
(b) The effect of silt from coal mines.
-------
123
The normal saprobic system analysis can be used for (a) but not for (b). The effect of silt is
recognized only by the following: when the effect is large, only few (or even no) species and only
a few individuals will be found on the bottom of the river, and when the effect is small, more
species and more individuals will be found. But in the case of organic pollution, we have indicators
for every class of water quality.
In this river, the pollution by the pulp mills (and paper mill) is the most serious problem,
especially the pollution by the factory of the Kokusaku Pulp Mill Company in Asahikawa City.
At Sunagawa. 90 km downstream from the factory the water color and BOD of the river water
have recovered, but COD value is still high (40 ppm), and almost the same value continues to the
estuary (Nakamura, 1961).
The importance of the dissolved organic materials from the peat zone in the drainage basin of the
I shikari should be also considered. These dissolved peat-originated substances are like the
substances from the pulp wastes (lignin, cellulose, etc.) - very hard for bacteria to decompose. It
must also be considered that the water temperature is low in most months of the year in this part
of Japan. These are the reasons why the brown water color and high COD value hold in the whole
middle and lower courses of the Ishikari River.
Suspended solids such as silt and sand from the coal mines are deposited on the bed of the Ishikari
River, to which they are carried by the tributaries. The silt and sand will be carried farther
downstream rolling on the bed. The bottom substances do not rest, since the bottom is not stable,
thus the development of the benthic microorganisms, including bacteria, is very poor, indicating
the reduced potentiality of natural purification of this river.
For the past two or three years the situation has improved. Many coal mines have stopped their
operations owing to the increased cost-gain relationship. The pulp mill in Asahigawa City is moving
part of its machinery to its other factory located nearer the sea.
REFERENCES
Kubo, T. and Kosaka, J. (1960) Influence of industrial wastes discharged to the Ishikari River
(between Asahigawa and Takigawa) on the distribution of organisms, Suisan-zooshoku-shiryo,
P. 1-
Kubo, T., Kosaka, J., Inoue, S.. Ito, T. and Yoshizumi, K. (1961) Influence of industrial wastes
discharged to the middle reach of the Ishikari River on the distribution of the aquatic
organisms, Suisan-zooshoku-shiryo, p. 1.
Nakamura, T. (1960) Studies on River Pollution: I. Studies on the water quality and
contamination of the Ishikari River, Report of the Hokkaido Institute of the Public Health,
No. 6.
Scheuring, L. and Zehender, C. (1961) Untersuchungen zur Stoffwechselphysiologie des
"Abwasserpilzes" Fusarium aquaeductum, Lagh. Schwez., Z. Hydrol., 24, 158.
-------
124
Tamura, T., Fuji, A. and Yuki R. (1961) Influence of industrial wastes discharged to the middle
reach (between Takigawa and Ebetsu) of the Ishikari River on the aquatic organisms,
Suisan-zooshoku-shiryo, p. 1.
-------
CHEMICAL EFFECTS OF SALMON DECOMPOSITION
ON AQUATIC ECOSYSTEMS
David C. Brickell and John J. Goering
INTRODUCTION
The Pacific salmon (Oncorhynchus sp.) is a valuable source of food energy from the sea and its
potential is not yet fully realized. The two most highly evolved species of the genus, O. gorbuscha
and O. keta, are true marvels of efficiency from the anthropomorphic vista. Upon hatching in the
gravels of coastal streams and rivers each spring, the salmon fry of these two species migrate
immediately to the sea to harvest the productivity of the ocean waters. Their migrations extend
over vast areas of the Central North Pacific.
After one year of residence in the ocean these salmon are compelled by instinct to return to their
natal waters. This homeward migration results in the transfer of millions of pounds of organic
matter, once widely distributed in the open ocean, to the coastal waters where salmon spawn.
Thus, the vast primary productivity of the open ocean, which is currently too diffuse to harvest
economically, is brought within man's grasp. This directed biomass transfer is sufficiently large to
be considered a "biological current" originating in the Central North Pacific and terminating in the
freshwater streams of the coast.
In addition to the large commercial harvest, of which a portion, about 20% by weight, is returned
to the ocean as waste from processing plants, millions more salmon enter freshwater streams to
spawn and inevitably to die and decompose. The fate and distribution of this organic matter has
previously received little attention.
Although the decomposition of salmon carcasses in the marine environment is a natural
phenomenon, the possibility that localized accumulations of seafood waste products, as might
occur in the vicinity of processing plants, may stress the marine environment beyond its capacity
to maintain stability must be considered.
Our interest in the fate of organic matter associated with seafood processing waste was stimulated
by a Bering Sea cruise of the R/V ACONA, cruise number 066, in June 1968. In samples taken off
the coast of Alaska in areas of significant seafood processing, namely in the area of Kodiak, Alaska
and Unalaska, Alaska, we observed a water layer with extraordinarily high concentrations of
ammonium. Near the processing plants toxic ammonium concentrations approaching 25 Mg-atom
NH^-N/liter were observed while lower yet distinctly elevated concentrations were observed over
tens of square miles of ocean in the area.
Other cruises of the R/V ACONA in the fjords of Southeast Alaska have revealed moderately high
concentrations of ammonium at about 20-30 m. Evidence suggests that the origin of this
ammonium is within the estuary rather than from the sea. Since the streams of Southeastern
Alaska are abundant with salmon, the observed concentrations of ammonium may represent the
results of carcass decomposition within the estuary.
125
-------
126
To increase our knowledge of the biological and chemical effects of the decomposition of seafood
material, we have initiated a study of the decomposition of salmon carcasses in a natural system in
southeastern Alaska (i.e. Little Port Walter estuary). The Pacific salmon migrates through this
estuary when returning to its natal stream to spawn. Following spawning, the fish "die and the
carcasses are eventually carried to the estuary where they sink to the bottom. During periods of
low stream flow, the dead carcasses may remain in the stream itself until higher stream flows
transport them to the estuary. In years of large escapements, the density of fish in the spawning
stream can be very high. In the system chosen for our work, spawning densities greater than six
fish per m2 have been recorded, although at the time the current study was conducted the
spawning density was slightly more than two fish per m2. Since our system involves primarily pink
salmon (O. gorbuscha). the average weight of the fish can be assumed to be 2-3 kilograms.
Thus, our study is concerned with the fate and distribution of some 75 metric tons of organic
matter in the form of salmon carcasses in one small estuary in Southeastern Alaska. We are
particularly interested in determining: (1) the effects of the salmon carcass decomposition on the
nitrogen chemistry of the water in which the decomposition occurs; (2) the form and distribution
of the organic matter which is returned to the marine system; and (3) the rate at which
remineralization occurs. This paper presents the results of our initial investigations.
METHODS
The system selected for study included a pink salmon spawning stream, Sashin Creek, and its
associated estuary. Little Port Walter, on Baranof Island, Southeastern Alaska. Another small
estuary in the vicinity, Toledo Harbor, was used as a control for the estuarine studies as it did not
support a salmon run.
Sashin Creek flows some 3,000 m from Sashin Lake to the Little Port Walter estuary. A high
waterfall prevents further upstream migration of the salmon, and hence spawning is limited to the
lower 1,200 m. The stream area above the water fall was used as a control for the stream studies
since this area revealed seasonal variations in water chemistry but was not influenced by salmon. A
permanent weir for counting fish entering the stream is located at high tide level.
The U.S. Bureau of Commercial Fisheries has maintained a research station at Little Port Walter
since 1934 for the purpose of collecting information on the freshwater ecology of pink salmon.
These studies have resulted in the definition of 3 distinct ecological areas of the stream: an upper,
middle and lower area. To maintain integrity with existing data, we observed the same sampling
boundaries in the stream. The total area of Sashin Creek available for spawning is about 13,600
m2. The width of the stream varies between 12 and 24 m. Stream discharge data for Sashin Creek
has been collected since 1951 and ranges from 8.0 to 700 cubic feet per second with a mean of 80
c.f.s.
The Little Port Walter estuary is small, consisting of an inner and outer bay with a constriction
between the two. The distance from the mouth of Sashin Creek at high tide to the entrance to the
outer bay is approximately 1.5 km, and the maximum width is about 0.4 km. The maximum
depth at low tide is 44 m for the outer bay and 21 m for the inner bay. The connecting channel is
27 m wide and about 6 m deep at low tide. Tidal variations range to 4.6 m. A detailed description
of the oceanography of Little Port Walter estuary is given by Powers (1962).
-------
127
Toledo Harbor, about 3 km south of Little Port Walter, has dimensions approximately the same as
the inner bay of Little Port Walter. The freshwater stream entering the harbor has a waterfall just
above high tide level and does not therefore support a salmon run.
Stream Studies
Prior to the arrival of the salmon, surface water samples were collected weekly in Sashin Creek
above the waterfall and at the lower end of each of the three designated ecological areas of the
stream. Analyses for NO:-N, NOj-N, NH4*-N, dissolved organic -N and inorganic phosphate were
performed. Nitrite was determined by the Griess method as applied to seawater by Strickland and
Parsons (1965). Nitrate was determined by conversion to nitrite in a cadmium-mercury reduction
column based on a method by Grasshoff (1964) and determined as nitrite. Ammonium was
measured using the recent method of Solorzano (1969). This method is specific for ammonium
and does not measure labile amino nitrogen. The ammonium method of Richards and Kletsch
(1964) was used to obtain the Bering Sea ammonium values. This method is not specific for
ammonium -N but measures a considerable fraction of labile amino nitrogen as well. Dissolved
organic nitrogen was determined by converting it to nitrate by ultraviolet light oxidation in the
presence of oxygen according to the method outlined by Armstrong, et al (1966). Reactive
phosphorus was determined by the method of Murphy and Riley (1962). Samples were collected
weekly beginning in August and continuing into November, about 2 months beyond spawning.
53'55'
O STATION LOCATIONS
IN DUTCH HARBOR
* SEAFOOD PROCESSING
SITES
BERING
SEA
APPARENT
TIDAL CURRENT
DIRECTION
UN4LASKA ISLAND
•J CAPTAINS O
BAY
166*35' I66'30'
FIGURE 1 Iliuliuk Bay, Unalaska Island, Alaska
-------
128
TABLE 1
Chemical and Physical Characteristics of Water in Iliuliuk Bay at Stations 2327
Near Seafood Processing Sites and at Station 2351 further Seaward
2349
Station
Number and Depth in Temp.
Location Meters C
2349
53°51.8!N
166 33.3 W
2347
53°52.8*N
166 32.7 W
2341
53°52.8'N
166 31.5 W
2337
53°53.2*N
166 31.1 W
2331
53°53.?!N
166 30.6 W
2327
53°54.2jN
166 30.2W
2351
54°01.3!N
166 04.5 W
0
5
10
15
20
30
40
0
5
10
0
5
10
15
25
0
5
10
15
25
0
5
10
15
25
0
5
10
15
0
5
10
20
30
50
5.70
5.63
5.01
4.81
4.34
3.70
3.30
7.35
5.52
5.20
6.71
5.98
5.39
5.03
3.83
5.95
5.86
5.47
5.19
3.79
6.70
5.87
4.92
4.64
4.11
6.02
5.68
5.22
-
4.87
4.88
4.88
4.87
4.88
4.87
. Salinity
o loo
32.051
32.005
32.237
32.266
32.294
32.360
32.343
24.906
32.171
32.223
30.782
31.667
32.202
32.366
32.373
31.788
31.820
32.265
32.313
32.381
30.781
32.109
32.304
32.347
32.362
31.777
32.146
32.376
32.391
-
32.431
32.408
32.502
32.412
32.523
Oxygen PO;3-P NH4+-N
ppm Mg-atomAiter
7.91
7.97
7.67
6.90
7.14
6.84
5.60
6.78
7.17
7.39
7.98
8.05
7.82
7.35
4.30
8.07
8.08
8.10
7.83
4.84
7.88
8.34
7.51
7.36
5.51
7.93
7.97
7.59
7.28
-
-
-
-
-
1.08
1.10
1.35
1.44
1.69
2.12
2.45
1.22
1.78
1.50
0.51
0.87
1.21
1.47
5.03
0.84
0.88
0.98
1.20
4.43
0.48
0.83
1.42
1.80
3.37
0.70
0.96
1.13
1.42
1.30
1.29
1.25
1.26
1.25
1.26
0.4
0.4
0.8
0.8
1.0
0.9
1.0
5.2
8.1
5.7
1.8
3.4
4.6
5.9
23.8
2.4
2.3
2.2
4.9
19.9
1.7
2.0
6.4
9.3
18.4
2.7
3.2
5.0
6.5
3.4
3.6
3.2
3.3
3.2
3.2
NO2~-N
/ug-atom/liter
0.17
0.16
0.19
0.19
0.20
0.20
0.22
0.19
0.26
0.23
0.06
0.11
0.15
0.16
0.33
0.12
0.14
0.12
0.15
0.33
0.06
0.12
0.17
0.21
0.28
0.09
0.10
0.13
0.16
0.16
0.17
0.18
0.17
0.17
0.17
NO3"-N
Mg-atom/liter
5.8
6.0
7.8
11.2
13.1
18.3
20.9
0.9
5.1
6.6
0.2
2.2
4.7
6.4
9.2
3.0
3.0
4.5
5.6
9.6
0.5
3.4
6.7
7.7
9.9
2.3
3.2
5.7
7.0
12.0
11.7
11.5
11.8
11.8
12.3
-------
129
Estuary Studies
In the estuary, weekly samples were collected at selected depths in both Little Port Walter and
Toledo Harbor. The same chemical analyses and procedures were employed as were used in the
stream studies.
RESULTS
Iliuliuk Bay and Bering Sea Studies
The locations of sampling stations and seafood processing plants on Amaknak and Unalaska
Islands, which are located near Unimak Pass in the Aleutian chain, are shown in Figure 1. The
oceanographic parameters observed in the area on cruise 066 indicated that during our sampling
period in June 1968 the tidal current entered the area from the northwest and exited toward the
northeast. The temperature-salinity-oxygen characteristics in the Captain's Bay area where station
2349 is located suggested an open ocean environment little affected by the waste of the processing
plants, while those stations to the northeast of the processing plants revealed rather marked
chemical alteration. Table 1 presents the chemical and physical parameters obtained at all stations.
Station 2349, indicative of the water entering the area of seafood processing, had a normal open
ocean ammonium concentration of approximately 1/ug-atom NH^-N/liter at 25 m while the
ammonium concentration at 25 m at station 2341 was 23.8jug-atom NH^-N/liter, which is an
extraordinarily high concentration for sea water. Station 2351, further seaward, exhibited the
chemical characteristics which result after the water from Iliuliuk Bay has mixed with Bering Sea
water. The ammonium values obtained in this study of Iliuliuk Bay were determined using the
method of Richards and Kletsch (1964). This method does not distinguish ammonium -N from
labile amino -N. Because of the highly organic nature of the seafood waste a considerable fraction
of the nitrogen reported here as ammonium -N could be amino -N.
Figure 2 presents a plot of the ammonium and oxygen concentrations with depths at stations
sampled in Iliuliuk Bay. The results show a drastic increase in the ammonium concentration of the
»2J49
2349
jj g-dtoms N*C-N/liter
FIGURE 2 NH4-N and oxygen at various stations and depths in Captain's Bay and Iliuliuk Bay
-------
130
waters while in residence in tliuliuk Bay. Whereas that water entering the bay from Captains Bay
has an ammonium concentration of approximately 1/ig-atom NH^-N/liter. The concentration of
ammonium in the bottom waters within the bay increases to almost 25/ug-atom NH^-N/liter.
Likewise, the oxygen concentration of the bottom waters is lowered. The ammonium
concentrations are high at intermediate depths, 5 - 10/wj-atom NH^-N/liter at 15 m, but the most
significant increase in ammonium and decrease in oxygen occurs between 15 and 25 m. Since the
25 m sample is near the bottom, the substantial increase in ammonium at this depth probably
results from decomposition of organic matter which has accumulated on the bottom.
The data for station 2351 indicated a well mixed water column with the ammonium distributed
rather uniformly through the column from the surface to 50 meters. Other seaward stations
displayed decreasing but elevated ammonium concentrations, suggesting that ammonium
originating in Iliuliuk Bay has an influence on the nitrogen economy of the surrounding ocean.
Stream Studies
After evaluating the data from the Iliuliuk Bay and Bering Sea study we decided to examine a
natural phenomenon which results in the accumulation of organic matter in the marine
environment similar to the situation observed in the vicinity of seafood processing sites. Salmon
inevitably die following spawning. This results in the accumulation of salmon carcasses both in the
freshwater stream where spawning occurs and in the receiving estuary. This natural system was
selected for study because it represented a cyclic phenomenon which could be followed from
beginning to end with little external influence.
About 30,000 pink salmon spawned in the 1200 m of Sashin Creek in 1969. After spawning and
death many of these carcasses remained in the stream for a period of time before being washed
into the estuary. Some of these were scavenged, but the majority of the carcasses were deposited
on the bottom of tine Little Port Walter estuary. In following the chemistry of salmon carcass
decomposition we were particularly interested in the nitrogen chemistry. Since fish flesh contains
much protein we felt that the various forms of nitrogen would be excellent indices of the rates of
biological decomposition of salmon carcasses.
Surface water samples were collected and analyzed weekly from four different sites in the stream;
above the waterfall which prevented further upstream migration of the spawners. Area O, and at
the lower boundary of each of the three defined ecological areas. These boundaries were as
follows: Area I, waterfall - 430 m downstream; Area II. 430 - 730 m downstream; and Area III,
700 - 1200 m downstream. The first 300 m of stream below the waterfall is not favorable for
spawning and hence only a few carcasses are deposited here.
Since the water at each sampling area downstream had been exposed to more carcasses, we
expected a systematic increase in the concentrations of metabolites resulting from carcass
decomposition as we moved progressively downstream.
Water samples collected prior to entrance of the fish into the stream showed little variationjn
water chemistry from above the falls to the mouth of the stream. On August 31, the NH4-N
concentration above the fall was 0.32 /Ltg-atom N/liter while at the stream mouth the concentration
was 0.46 Mg-atom N/liter. Similarly, the dissolved organic nitrogen concentration above the falls on
that date was 3.3 MQ-atom N/liter, while at the stream mouth an organic nitrogen concentration of
3.8 Mg-atom N/liter was observed.
-------
131
Spawning activity commenced in late August and by early September dead carcasses began to
appear in the stream. The first noticeable chemical effects of decomposition were observed in the
stream on September 14. On this date, the ammonium concentration above the falls was 0.40
jug-atom N/liter while at 1200 m downstream the concentration increased to 1.73 jug-atom N/liter.
Likewise the dissolved organic nitrogen concentration increased from 4.0 jug-atom N/liter to 7.4
jug-atom N/liter during the traverse. As the number of carcasses increased, the downstream increase
in ammonium nitrogen and dissolved organic nitrogen became greater. On September 29 we
observed the greatest difference in stream chemistry between the control area and the lower end of
the stream. The ammonium concentration increased from 1.66 jug-atom N/liter above the falls to
7.80 /ug-atom N/liter at 1200 m downstream. Dissolved organic nitrogen increased from 5.7
/;g-atom N/liter to 17.7 jug-atom N/liter. The concentration of NH4+-N and dissolved organic -N
observed at each sampling site and date are presented in Table 2 and in Figures 3 and 4.
TABLE 2
The Concentrations of Ammonium -N and Dissolved Organic -N (D.O.N.)
Observed in each Sampling Area of Sashin Creek
Concentrations are Expressed as jug-atom-N/liter
A AreaO
Date
8/24/69
8/31/69
9/7/69
9/14/69
9/22/69
9/29/69
10/8/69
10/16/69
10/28/69
NH4-N
0.20
0.32
0.32
0.40
1.31
1.66
1.56
1.44
1.63
DON
3.04
3.29
3.38
4.04
5.83
5.71
5.23
5.22
4.84
Area I
NH4-N
0.60
0.38
0.64
0.33
1.44
4.59
4.48
4.04
2.46
DON
3.18
3.39
3.64
5.31
6.44
7.38
6.72
6.11
6.02
Area II
NH4-N
0.62
0.38
0.92
1.40
3.22
6.55
6.62
6.72
3.88
DON
3.27
3.49
4.84
6.10
8.22
12.40
10.52
8.43
7.02
Area
jT
NH4-N
0.68
0.46
0.98
1.73
4.17
7.80
7.66
7.02
5.63
III
DON
3.67
3.84
5.39
7.36
9.71
17.67
16.8
10.32
8.64
Estuary Studies
Water samples in the Little Port Walter estuary and in the control estuary, Toledo Harbor, were
collected weekly at 1, 4, 8 and 12 m. Dissolved organic nitrogen proved to be the most valuable
index of the salmon carcass decomposition phenomenon. The dissolved organic nitrogen
concentrations observed in Little Port Walter and Toledo Harbor appear in Table 3. Figure 5
compares graphically the dissolved organic nitrogen of the surface waters of the two estuaries.
Figure 6 is a comparison of their organic nitrogen concentrations at 12m.
The first effects of salmon carcass decomposition on the water in Little Port Walter estuary, as
evidenced by increases in dissolved organic nitrogen, were not observed until about 15 days after
-------
132
8/24 8/31 9/7 9/W 9/22 9/29 IO/8 10/16 10/28
DATE
FIGURE 3 Concentrations of NHVN in Sashin Creek surface water
3/16/7O
increases were first observed in the stream. Heavy rains in late September resulted in flushing most
of the carcasses from the stream into the estuary where they sank to the bottom. The dissolved
organic nitrogen concentration in the surface water of Little Port Walter estuary began to increase
on September 22 and reached a peak of 103 jug-atom N/liter on October 28 (Fig. 5} while the
surface concentration in Toledo Harbor remained almost steady at about 5.4 pig-atom N/liter
throughout this period.
The data on ammonium concentrations for the two estuaries did not reflect the immediate effect
of carcass decomposition. The concentrations in fact declined during the fall months from a
summer high. The ammonium concentrations observed at 12 m depth in the two estuaries are
presented in Figure 7.
-------
133.
8/24
8/3!
9/7
9/14
1/8
10/16
9/22 9/29
DATE
FIGURE 4 Concentrations of dissolved organic -IM in Sashin Creek surface water
-------
134
TABLE 3
Dissolved Organic Nitrogen Concentrations Observed in Little Port Walter
and Toledo Harbor Estuaries in Summer and Fall of 1969
Concentrations Expressed asjug-atom-N/liter
Little Port Walter
Depth 8/24
m
1 2.42
4 2.64
8 2.42
12 2.56
Toledo Harbor
1 2.07
4 1.87
8 3.21
12 2.46
10 -
8/31 9/7 9/14 9/22
2.40 3.96 3.27 3.78
1.46 3.03 3.40 1.48
1.32 3.02 3.18 1.09
3.90 3.38 3.65 1.80
1.52 1.45 1.34 1.23
0.65 1.69 1.41 1.20
1.60 1.98 1.88 1.32
3.09 2.14 2.35 1.92
9/29 10/8 10/16 10/28
5.46 6.62 9.20 10.33
3.36 7.52 4.86 10.83
3.50 2.55 11.61 9.11
6.53 8.92 16.26 14.26
5.24 5.48 5.42 5.32
5.60 5.66 5.63 4.27
4.98 3.45 4.99 5.39
4.93 4.63 5.02 5.42
1 1 1
_/"\
11/14
5.62
4.83
7.68
12.43
5.02
4.22
4.87
5.20
_
J? LITTLE PORT WALTER / \
'— g . surface — >. / \
ai
o
-0 6 -
5
o
m
m
T> 4 -
E
o
0 ( 1
8/24
• sf
• ^
'A A A
\
\(
-^ 9 I TOLEDO HARBOR
/ surface
/
"»A A A J-
~-A A
1 1 1 1 \
8/31 9/7 9/14 9/22 9/29 10/8 10/16 10/28
DATE
-
11/14
FIGURE 5 Concentrations of dissolved organic -N in surface water in Little Port Walter and
Toledo Harbor estuaries
-------
18
14
Z
i
o>
•o
I
o
m
n
E
2
o
10
LITTLE PORT WALTER
IZ METERS
TOLEDO HARBOR
METERS
I
8/24
8/31
9/7
9/14
9/22
9/29
10/8
10/16
10/28
11/14
DATE
FIGURE 6 Concentrations of dissolved organic -N in 12m water in Little Port Walter and Toledo
Harbor estuaries
-------
136
LITTLE PORT WALTER (12m)
• TOLEDO HARBOR (12m)
O LITTLE PORT WALTER (surface)
O TOLEDO HARBOR (surface)
8/24 8/31 9/7 9/14 9/22 9/29 IO/8 10/16 10/28 11/14 3/16/70
DATE
FIGURE 7 Concentrations of ammonium -N at 12m in Little Port Walter and Toledo Harbor
estuaries
-------
137
DISCUSSION
The surprisingly large concentrations of NH^-N and the oxygen depletions observed in Iliuliuk Bay
indicate that the decomposition of seafood material in a restricted body of water can influence the
nitrogen chemistry of the estuarine environment. The organic waste matter resulting from seafood
processing plants located on Iliuliuk Bay is emptied into the estuary rather continuously. Thus the
highest observed concentrations of NH4*-N at station 2341, 23.8 jug-atom N/liter may well
represent the remineralization of organic material that has accumulated over a long period of time.
It would appear that such extremely high concentrations of ammonium are potentially toxic to
fish (Ball, 1967} and to other forms of life. Other effects of decomposition such as oxygen
depletion are also probably detrimental to the organisms residing in the bay.
It is important to determine whether elevated concentrations of ammonium as were observed in
Iliuliuk Bay occur naturally in the marine environment. It appeared from the knowledge available
on the nitrogen chemistry of the sea that it would be unlikely that such high concentrations of
NH^-N would occur naturally, and that the observed situation was therefore the result of
industrial stress on the environment. We felt that the phenomenon of salmon carcass
decomposition most closely duplicated the industrial situation, and we were therefore interested in
determining what level of nitrogen compounds were present in systems where salmon were
decomposing.
It is difficult to compare the Richards and Kletsch (1964) ammonium values for Iliuliuk Bay to
those obtained by the Solorzano (1969) technique in Little Port Walter, since it is uncertain what
fraction of the organic nitrogen was recorded by the Richards and Kletsch method as ammonium.
Our investigations in Little Port Walter suggest that dissolved organic nitrogen is the initial
decomposition product and organic nitrogen (i.e., labile amino-N) may have contributed heavily to
the ammonium measured in Iliuliuk Bay. If the ammonium -N and dissolved organic -N for Little
Port Walter are summed, a combined dissolved organic -N - ammonium -N concentration of 18.2
jig-atom N/liter occurred near the bottom on 10/16/69. This approximates the value of 23.8
fig-atom of NH^-N/liter observed in Iliuliuk Bay and indicates that high concentrations of
ammonium and organic nitrogen are not restricted to waters receiving waste from seafood
processing.
The ammonium and dissolved organic nitrogen concentrations observed in Sashin Creek indicate
that decomposing salmon carcasses have a significant influence on stream chemistry. It has been
observed that survival of salmon eggs in the Sashin Creek spawning beds was significantly lower in
the lower area of the stream. Area III, than in the upper areas. Several reasons have been presented
by McNeil (personal communication) to explain this phenomenon. The chemical parameter
normally associated with survival of eggs in the spawning bed is dissolved oxygen, but we fee I that
ammonium or other products of carcass decomposition might have a significant influence on egg
survival, especially when stream flow is low.
The large dissolved organic nitrogen concentrations observed in the surface waters of Little Port
Walter undoubtedly reflected the decomposition of salmon carcasses occurring in the stream and
in the intertidal zone. The bottom of the estuary received most of the carcasses and it is here that
final decomposition occurs. Sediment samples of the Little Port Walter estuary bottom showed a
gelatinous quality even for samples taken prior to the arrival of the fish, suggesting that the
sediment may act as a nutrient sink. This gelatinous nature probably results from the long term
accumulation of fish carcasses. All sediment samples from Toledo Harbor were granular.
-------
138
The apparent increase in the concentration of dissolved organic nitrogen observed in the control
estuary, Toledo Harbor, between September 22 and September 29 was probably the reflection of
heavy rains, wind and heavy cloud cover which resulted both in high run-off from, the land and
death of plankton in the water column.
It is evident that this brief study is only a beginning in our understanding of the fate of organic
matter deposited, naturally or by industry, in the marine environment/Our attempts to follow the
decomposition of salmon carcasses employing nitrogen chemistry was rewarding in that effects
were clearly demonstrable. Since nitrogen comprises only a small proportion of the organic matter,
and since remineralization rates are slow, it is evident that more sophisticated techniques must be
utilized to obtain a clear understanding of the process. Possibly, remineralization of the organic
matter resulting from carcass decomposition does not occur within the confined body of water
where the carcasses are deposited. The early stages of decomposition might result in the
production of soluble complex organic compounds which are transported from the local area
before remineralization occurs.
ACKNOWLEDGMENTS
This research was supported in part by Office of Water Reserach Grant B-015-ALAS and by the
National Science Foundation Grant GB-8636.
REFERENCES
Armstrong, F. A. J., Williams, P. M. and Strickland, T. D. H. (1966) Photo-oxidation of organic
matter in sea water by ultraviolet radiation, analytical and other applications. Nature,
211,481.
Ball, I. R. (1967) The relative susceptibilities of some species of freshwater fish to poisons, I,
Ammonia, Water Research, 1,587.
Grasshoff, K. (1964) Zur Bestimmung von Nitrate in Meer und Trinkwasser, Kiel. Meeresforsch,
20,5.
Murphy, J. and Riley, J. P. (1962) A modified single solution method for the determination of
phosphate in natural waters. Anal. Chim. Acta,, 27, 31.
Powers, C. F. (1962) Some aspects of the oceanography of Little Port Walter estuary, Baranof
Island, Alaska. Fish. Bull., 63, 143.
Richards, F. A. and Kletsch, R. A. (1964) The spectrophotometric determination of ammonia and
labile amino compounds in fresh and seawater by oxidation to nitrite, In Y. Miyake and T.
Koyama (ed.). Recent Researches In The Fields of Hydrosphere, Atmosphere, And Nuclear
Geochemistry, Maruzen O., Tokyo.
Soloranzo, L. (1969) Determination of ammonia in natural waters by the Phenolhypochlorite
method, Limnol. Oceanogr., 14,799.
Strickland, JJ. D. H. and Parsons, T. R. (1965) A manual of sea water analysis (with special
reference to the more common minor nutrients and to paniculate organic material). Bull.
Fish. Res. Bd., Can., 125,79.
-------
PHOSPHORUS BINDING MECHANISMS DURING
SELF-PURIFICATION OF POLLUTED LAKES
Jan Werner
INTRODUCTION
The discharge of sewage into lakes and rivers gives rise to different and fundamental changes.
Among the various parameters of importance that can influence the ecosystems, the following
three will be discussed:
1. The oxygen consuming substances
2. The phosphorus compounds
3. The alkalinity
The increases in phosphorus concentration and alkalinity are two factors which are supposed to
affect the primary production of a limnic system directly and positively (Christie, 1968; Mortimer
1941, 1942). The increased production will, in combination with the external input of organic
substances, sometimes cause a complete depletion of the oxygen in the bottom waters during
periods of stratification. The sediments will be especially exposed to such conditions, as the
organic particles will settle to the bottom and form part of the upper surface of the sediment.
In this connection the exchange reactions between water and sediments seem to be of considerable
importance. The sediments act as large buffers and storing places for various types of solid
materials. Exchange mechanisms wilt cause a shift of certain substances between the solid and the
dissolved states. Phosphorus is one of the substances that will be affected by the exchange
mechanisms. The fundamental experiments of Mortimer {1941, 1942) corroborate the general
experience that when oxygen is available to the sediments the transfer of phosphorus from a solid
to a dissolved state will be very limited. There will on the other hand be an increase of soluble
phosphorus in the sediment system when oxygen is depleted.
CIRCULATION OF PHOSPHORUS
AND ITS ACCUMULATION IN THE SEDIMENT SYSTEM
During summer, process 2 tends to dominate over process 3 with an enrichment of phosphorus in
the hypolimnion and in the sediment system as a consequence. During winter, process 2 will again
dominate over process 3, especially during the period following the formation of an ice cover. The
spring and fall turnovers will on the other hand cause an even distribution of the phosphorus
throughout the whole water mass, i.e. process 3 will be favored over process 2.
The sediment material with which phosphorus is associated will be broken down successively to
more simple substances by biochemical processes. Whether phosphorus is bound to organic
molecules or not, orthophosphate will be the principal end product. To reduce the amount of
phosphorus circulating within a body of water, mechanisms causing an accumulation in the
sediment must be made to operate. There are in the first instance two principally different
mechanisms possible.
139
-------
140
Air
Epilimnion
Hypolimnion
/&S/SS/S//////S///S///S////S/S//////S//S//SS//SSSSSS
Sediment
I. Dissolved ©—»© in biomass
2. P in biomass—»sedimerting ©
3. Sedimenting ©—^dissolved ©
FIGURE 1 Circulation of phosphorus and its accumulation in the sediment system
1. An essential part of the settling organic phosphorus compounds will decompose slowly. The
accumulation according to this mechanism will reflect a dynamic equilibrium determined by the
quantity settling, the rate of decomposition, and finally the quantity released from the sediment.
2. Phosphorus compounds, especially orthophosphate, will be bound to undefined inorganic
solids in the sediment system. There will be set up a chemical equilibrium between absorbed
phosphorus and phosphorus dissolved in the interstitial water and in the water above and near the
sediment surface.
BINDING OF PHOSPHATE TO FERRIC IRON
In a lake with low primary production and with a limited input of oxygen consuming substances,
oxygen will normally be available in the bottom waters. In such a system any ferrous iron will be
oxidized to ferric iron, provjded it is not present as a very stable chelate. Ferric iron forms stable
and partially soluble molecular aggregates under the prevailing conditions. These aggregates may
have different anions incorporated, of which the following three are of special interest: phosphate,
hydroxyl and humic substances. Iron is normally present in a large excess compared to
phosphorus. The conditions in the sediment may be illustrated as in Figures 2 and 3. If ferric
hydroxide in the sediment is in excess the solubility of the ferric iron will be determined by the
solubility product of ferric hydroxide. We may assume that the hydroxide exists in various loosely
crystallized modifications. The lack of defined crystallinity makes it fairly reactive. The reported
stability constants for ferric hydroxide range from 10~^7 to 10"44 (Feitknecht, 1959). This
reflects that the stability of ferric hydroxide will depend on the conditions at which it is formed
and also its age, etc. This raises difficulty when a value relevant to the conditions in a sediment
system is to be selected. It seems reasonable, however, that the stability constant would be closer
to 10 than to 10 , i.e. the hydroxide should be comparatively reactive.
If ferric hydroxide is in excess compared to ferric phosphate, the concentration of soluble
phosphate in bottom waters and its dependence on pH may be illustrated as in Figure 4. The
temperature of the bottom water has been assumed to be about 5° C. The ionic product of water,
K>A., will then be 10 . The derived figure should be considered as an approximation as the
W
-------
141
Fe(+m) humic substances
Fe(+IE) hydroxide
FeMII) phosphate
FIGURE 2 Binding of phosphate to ferric iron
1-
cr
it
0.15
0.10
0.05
0
i
Bornsjon
(oligotrophic)
Mg
028
•
-
Fe
Al
Co
Fe
1 C
PtoT°
^tot
PM
PC
>4
Albysjdn
(eutrophic)
Efi
n
Fe_|n^
ptot"
Sir167
Mg
Ca
k
r-%
Lillsjdn
(hypertrophic)
Pe_Qp
Ptot
Fe
Al
^H*.
^'=24.6
Mg
I [^£04
FIGURE 3 Sediment analyses (Average 0 - 6 cm)
-------
142
3
4
5
6
8
9
K)h
12
^(total P04)
M
eight analyses, all below
detection limit
lower _
detection limit
678
• = calculated values
o = experimental values
FIGURE 4
IO
II
corrections of certain data for temperature have not been possible. The constants used for
calculation of the concentrations of different ionic species of phosphate have been taken from
Egan et al. (1961). The figure also includes a curve derived from an actual experiment. It appears
that the experimental curve fits fairly well to the theoretically derived curve when 10 is
selected as stability constant for ferric hydroxide under the discussed conditions.
The figure indicates that:
a) ferric hydroxide in a sediment system is of a relatively reactive type, and
b) the concentration of soluble phosphate is highly dependent on pH, being near or below the
detection limit under the condition expected in the system.
BINDING OF PHOSPHATE TO Ca2*, Mg2* AND Al3*
Phosphate ions can be bound to ferric iron, aluminum, calcium and magnesium. Precipitation of
phosphate by calcium or magnesium takes place only at higher concentrations than those
discussed. At the pH which prevails in the bottom waters, precipitate of calcium or magnesium
phosphate takes place only to a limited extent.
Aluminum will on the other hand react with phosphate in a way analogous to ferric iron. The
resulting complexes have approximately the same stabilities and the same dependances on pH. As a
consequence, any discussion of the role of iron as a phosphate-fixing agent in sediments must
consider the possible competition between aluminum and ferric iron. This is of special importance
under conditions when ferric iron might be reduced to the ferrous state.
-------
143
The total amount of iron and aluminum present at the surface of the sediment exceeds the
phosphorus content on a molar basis more than tenfold. The fact that in spite of this a complete
fixation of phosphate in the sediment system does not take place will need careful analysis.
REDUCTION OF FERRIC IRON AND OXYGEN
The reduction of ferric iron must be considered with reference to the following:
a) Fe3* is an important oxidizing agent and a redox buffer at the surface of the sediment.
b) When ferric iron is reduced to ferrous iron, it loses its properties to bind phosphate, hydroxyl
ions and humic substances. A physico-chemical analysis of the reaction
Fe^+e" Z Fe2* E° = 770 mVat 25°C (Schumbet al., 1937)
will give a redox value according to Nernst's equation
(Fe3*!
EFe3*/Fe2* = 770+ 55 '°9 ^<55 mV 3t 5°C)
The Fe^-concentration will be determined by the reaction
Fe3*+3OH" ? Fe (OH)3 (s)
Fe3* will vary according to the expression
10-37.0
(OH')3
ction of water being 10"
with pH according to
With the ionic production of water being 10' at 5° C, the ferric ion concentration will vary
(Fe3*) = 10'37>0+ 3-14.7 - 3 pH = 10*8'6 ' 3 pH
If the concentration of Fe2* is assumed to be 10 M (0.6 ppm) and if we use the expressions for
Fe3* and Fe2* given by Nernst's equation the following expression will result
1(f8.6-3pH
EFe3*/Fe2+= 77° * 55 log
10'5
Epe3»ype2+ as a function of pH is given in Figure 5.
Oxygen may be reduced in the following way:
1/2O2+H2O+2e" £ 2 OH' E° = 401 mVat 25°C (Latimer, 1952)
The redox potential varies according to
55 PO-,
E02/OH' = 401+—log
-------
144
The 02/OH" system will be the determinant for the oxidation or reduction of iron.
PQ is given as the partial pressure of oxygen. The oxygen content in the atmosjahere is 21% of
volume giving PQ = 0.21 atm. At saturation in fresh water and at 5° C this corresponds to an
oxygen concentration of 12.5 ppm. Henry's law gives
0.21 = K -12.5 K = 0.0168
Po = 0.0168 (O2 ppm)
We introduce this expression into the above Nernstian expression and obtain
55 (02 ppm)
See Figure 5
At a low oxygen concentration, e.g. 0.1 ppm, EQ /QH~ varies in the way indicated in Figure 5.
From the values given in this table the following conclusions can be drawn:
a) Also, at as low an oxygen concentration as 0.1 ppm the resulting redox potential will cause
ferrous iron to be oxidized to ferric iron.
b) The ferric hydroxide will be more stable at higher pH than at lower pH, as is seen from the
difference between the two redox systems shown in Figure 5.
56789
:• = 770*55 log tj£*r] ; [Fe3*] = I06>l
[021 in ppm
FIGURE 5
10 p
ft
-------
145
CHEMICAL DIFFERENCES BETWEEN FERRIC AND FERROUS IRON
It might be of value to use the generalizations of Ahrland et al. (1958) who have outlined some
fundamental differences between hard and soft metal acids. The hard metal acids tend to be in a
high oxidative state and have their electron shells tightly bound to the atomic nucleus. The
electrons are less able to polarize and the chemical bonds formed by these ions and their
counterions show a high degree of ionic bonding. Typical hard acids are Mg2+, Ca2+, AI3+ and Fe3*.
The H* ion also shows the characteristics of a hard metal acid.
The soft metal acids are characterized by no oxidative state, or a low one. The electron shells are
larger and more polarized and the chemical bonds also tend to be more polarized, leading to a high
degree of covalency. Typically soft acids are Cu+, Cd2+, Hg2+ and Hg2+ . Fe2*, Cu2+ and Zn2+ show
characteristics that place them in between the hard and the soft acids. In their relations to anions,
the hard and soft acids react differently. The following sequences of stabilities of complex ions are
often found (Fig. 6). The atoms in the figure are assumed to be incorporated in the active sites of
the anions.
Hard Metal Acid: N>P>As>Sb>Bi
Soft Metal Acid: NAs>Sb>Bi
Hard Metal Acid: 0>S>Se>Te
Soft Metal Acid: O<5~Se~Te
Hard Metal Acid: F>CI>Br>l
Soft Metal Acid: F
-------
146
Fe2*
^*P- poor
I. oxidation Fe2**02—»Fe3*
2. oraanic-P Hcttrlg»po4
3. Fe3** PQj—* FePQ. precipitate
4. nettodiffusion of P04
FIGURE?
water
oxidative
zone
reducing
zone
pH-DEPENDENCE OF FERROUS PRECIPITATES
Any tendency to form insoluble ferrous compounds during the establishment of anaerobic
conditions in the sediment will cause the oxidation - from a kinetic point of view - to regress to
ferric iron during instances when aerobic conditions are reestablished. The carbonate and sulfide
are important in this connection. The solubility product of FeC03 is given by
(Fe2+)(CO|') = 10't0'7
The carbonate is in equilibrium with hydrogen carbonate and carbonic acid. The solubility of
FeCO3 will thus be a function of pH as shown in Figure 7 (Weber, Stumm, 1963).
FeS reacts in the same way, the solubility of which is
(Fe2+)(S2-)=10-19'3
The sulfide is in equilibrium with mono- and dihydrogen sulfide. The solubility of ferrous sulfide
will vary with pH as is shown in Figure 8 (Hutchinson, 1957).
The above stresses the fact that the stabilities of all major ferrous compounds that are present in a
sediment system will be functions of pH. It may be assumed that the combination of ferrous ions
with any organic matter present in the sediment will follow the same general rule.
The following will take into consideration how the alkalinity of the total lake system will
influence the equilibria outlined above.
-------
147
ROLE OF HYDROGEN CARBONATE AND CARBONIC ACID
The pH of a lake system is influenced by a variety of factors. These include biochemical processes
and exchange reactions between the atmosphere and the water and between the water and the
sediment. The carbonate system will normally cover around 90% of the total buffer capacity.
Organic matter of different origin will cover the remaining part. pH varies around pH 7 according
to the relation
{HCO-3)
pH = 6.52+ log
(H2C03)
at 5° C
Discharge of sewage will increase not only the phosphorus concentration but in addition the
concentration of bases. These bases are transformed into hydrogen carbonate by biochemical
processes either in a treatment plant or in the receiving water. The increase in hydrogen carbonate
concentration will raise the pH according to the above expression. A higher pH of the total mass
will thus result. The change in alkalinity will influence both the trophogenic zone and the
0
2
4
6
8
10
2 4 6 8 10 12 pH
Solubility of Fe2* = f(pH) (10) in l(T3M C03
0
2
4
6
8
10.
H2Sfofal -ICPM
2 4 6 8
Solubility of Fe2* = f(pH) in
FIGURE 8
10 12 pH
H2S
-------
148
sediment system. The influence of pH on the redox potential of the 02/OH" and the Fe3+/Fe2+
systems, upon the binding of phosphate to Fe3* and to AI3+ and upon the solubility of various
Fe2+- precipitates has been mentioned previously. Moreover, all the trace elements will probably
act in a way analogous to Fe2* by being less soluble at higher pH.
The theory discussed in this paper illustrates that in order to enhance the capacity of the
sediments to act as a sink for phosphorus, advantage could be taken of the fact that the binding of
phosphate to ferric iron and to aluminum will increase in stability when the pH is lowered. The
validity of this was tested previously for the total organic production in a series of experiments
where the alkalinity of a nutrient-rich water was changed by the addition of controlled quantities
of acid (Werner, 1969). The addition of acid-reduced hydrogen carbonate by partially replacing it
with CI" or SO|*. The results obtained clearly indicated that the primary production depends on
the alkalinity and the pH, among other factors. Lowering of the alkalinity was accompanied by an
approximately linear decrease of production, measured as the total amount of organic carbon. The
following will describe supplemental experiments where the reactions of a sediment system to
changes in the hydrogen carbonate concentration and the pH have been examined.
EXPERIMENTAL
Sediment and water were taken from the hypertrophic lake Lillsjon near Stockholm. The lake was
heavily polluted by municipal sewage. The sediment was taken from the bottom by an Ekman
dredge and the water was taken with a Ruttner sampler. The time of sampling, 6th of October,
coincided with the fall turnover. 19% supersaturation, combined with a high pH, indicated that
biological production was still taking place. Table I gives the analytical data of the sediment and
the water sample.
50 ml of sediment and 450 ml of water were added to each of 24 - 500 ml Erlenmeyer flasks. The
flasks were divided into six series and known amounts of 0.1000 M HCI were added to each group.
The flasks were shaken slightly and kept at room temperature. Illumination was made so as to
simulate the light conditions during fall circulation. The added amounts of HCI are shown in Table
TABLE!
Analysis of:
Sediment
suspended matter 50.3 g/1
total organic carbon 1.4 g/1
total iron 1.0 g/1
total phosphorus 0.108 g/1
Water
dissolved oxygen 13.7 mg/1
temperature 8.8° C
pH 9.1
alkalinity (HCO,) 1.58 mekv/1
total organic carbon 17 mg/1
total iron 0.120 mg/1
total phosphorus . 0.500 mg/1
phosphate 0.235 mg/1
-------
149
2. Visual inspection showed that the suspended matter settled much faster in the flasks with the
largest amount of acid added: series 1 and 2 more than series 3 to 6. There was a difference seen
within series 3 to 6, the matter in series 6 settling more slowly than in series 5, etc. This is what
was expected. The added hydrogen ions neutralized excessive negative charges, thus making the
repulsion between the particles less efficient. The particles agglomerated and settled more easily.
The aqueous phase was analyzed on day 11 (Table 2). Each value was the average of four
replicates. It appears from this table that the alkalinity has decreased somewhat irregularly. This
may be explained by the slow rate of certain reactions in which the solid phase takes part. Calcium
and magnesium have partly been replaced by hydrogen ions, thus appearing in the aqueous phase.
The iron is probably associated with colloidal or dissolved substances. Finally, the total
phosphorus concentration has decreased in all the flasks.
TABLE 2
Analysis of Water After 11 Days
Total
Series Added Acid Alkalinity Fe P Ca Mg
mekv/1 mekv/1 jzg/1 jug/1 mg/1 mg/1
1 1.80 0 550 286 24.4 6.6
2 1.44 0 600 288 21.2 6.05
3 1.08 0.30 600 245 18.9 5.80
4 0.72 0.52 620 243 17.4 5.48
5 0.36 0.95 680 242 15.2 5.35
6 0 1.55 740 267 14.4 5.35
The flasks were placed in the dark at 20° C and covered with a membrane. This limited oxygen
diffusion in the water, thus simulating the conditions in the near bottom water and the sediment
during a period of stagnation. The elevated temperature was selected in order to speed up certain
reactions. Oxygen and pH were analyzed after a total of seven days in the dark (Table 3).
The flasks were analyzed again after an additional period of 31 days in the dark under the same
conditions (Table 4).
The iron concentration had dropped considerably in all instances, and could be detected only in
the series having the alkalinity reduced to zero. In the series with a measurable alkalinity, the
alkalinity had decreased in all instances. One of the reactions that probably had taken place in the
sediment, causing the alkalinity to decrease, is the following:
FeS+9/4O2 + 2OH'+ 1/2 H20-" Fe (OH}3 + SOl'
The pH values in the extreme series 1 may very well be considered only of interest as illustrations
of the discussed theory, as may be those of series 2 and 3. The concentration of hydrogen
carbonate is zero in these series. The values for oxygen indicate that bacterial decomposition of
-------
150
TABLE 3
Analysis of Water After 7 Days In The Dark
02
Series pH mg/1
1 4.50 4.1
2 5.31 4.2
3 5.45 3.3
4 6.45 4.5
5 6.83 4.0
6 7.15 4.8
the organic* continues in all series. A consequence is that carbonic acid and carbon dioxide are
present in excess. In accordance with theory, the phosphorus concentration shows the largest drop
in the series having no alkalinity and a low pH. Also the series having a measurable alkalinity shows
an apparent relationship between the parameter and the phosphorus concentration. The lower the
pH of the water the less phosphorus will be in solution (Fig. 9), thus indicating that phosphorus
will be absorbed by the sediment system in a direct dependance upon pH when aerobic conditions
are maintained.
It is felt that the above discussion and the results reported may assist in the interpretation of the
phenomena associated with the discharge of sewage to natural waters of different types concerning
alkalinity and character of sediment system.
TABLE 4
Analysis of Water After 40 Days In The Dark
Total
Series Alkalinity
mekv/1
1 0
2 0
3 0
4 0.32
5 0.85
6 1.44
Fe
AtgA
170
60
30
trace
trace
trace
P
jug/1
10
16.1
24.4
79
92
112
P04-P
J"g/l
10
13.2
18.0
61
84
108
PH
4.1
4.8
5.3
6.1
6.6
7.0
02
mg/1
2.35
1.73
2.02
1.73
1.50
1.60
-------
• =
P04
total P
FIGURES
-------
152
REFERENCES
Ahrland, S.. Chatt, J.. Davies, N. R. and Quart, Rev. (1958) (London), The relative affinities of
ligand atoms for acceptor molecules and ions. Vol. 12, p 265.
Christie, A. E. (1968) The Ontario Water Resources Commission, Div of Research, Publ. No. 32.
Egan, E. P. Jr., Wakefield, Z. T. and Luft, B. B. (1961) Low temperature heat capacity, entropy
and heat of formation of crystalline and colloidal ferric phosphate dihydrate, J. Phys. Chem.,
65, 1265.
Feitknecht, W. (1959) Z. Electrochem., 63,1979.
Hutchinson, E. G. (1957) A Treatise On Limnology, Vol. I, 760.
Latimer, W. M. (1952) Oxidation Potentials, 2nd Edn, Prentice-Hall, New York.
Mortimer, C. H. (1941) The exchange of dissolved substances between mud and water in lakes, J.
Ecology. Vol. 29, p 280.
Mortimer. C. H. (1942) The exchange of dissolved substances between mud and water in lakes, J.
Ecology, Vol. 30 p 147.
Schumb, W. C., Sherrill, M. S. and Sweetser, S. B. (1937) The measurement of the molal
ferric-ferrous electrode potential, J. Am. Chem. Soc., 59,2360.
Volleuweider, R. A. (1968) Organization for Economic Cooperation and Development, Paris
DAS-CSI-68.27.
Weber, J. W.. Jr. and Stumm, W. (1963) Mechanism of hydrogen ion buffering in natural waters, J.
Am. Wat. Works. Assn., 55. 1553.
Werner. J. (1969) Havsforskarmotet i Lund, Paper No. 18.
-------
CRITICAL REVIEW OF PAPERS ON RECEIVING WATERS
Peter A. Krenkel
INTRODUCTION
Temperature has profound effects on biological, chemical, and physical processes. It is particularly
appropriate to investigate these effects under cold weather conditions, inasmuch as a paucity of
information exists regarding waste treatment and water quality management in the Arctic.
The organizers of this conference are to be commended in gathering together such an eminent
group of scientists and engineers, and it is a great honor to have the opportunity of discussing a
portion of the papers presented.
I have attempted to be as objective as possible in the following discussions, and it is hoped that the
authors involved will accept my comments as constructive discussion rather than criticism.
Synoptic Study of Accelerated Eutrophication in Lake Tahoe, California-Nevada
Charles Goldman
Lake Tahoe is probably one of the most beautiful lakes in the world and changes as described by
Dr. Goldman are quite obvious to me, since I spent many days there some 20 years ago. The real
problem is, of course, people.
Measurements included primary productivity of phytoplankton and periphyton, species
composition, biomass, and biotic diversity. I believe the productivity information is of particular
significance inasmuch as the methodology utilized appears to yield maximum information in a
minimum of time and effort.
The results of the study demonstrate the effects of the added nutrients on the receiving water and
certainly demonstrate the need for control measures.
The author is to be complimented on the analysis of his data. There is little doubt that the future
of water pollution control will dictate more quantitative work from the biologists, as is the case
with Dr. Goldman's work. The use of statistics and parameters, such as a species-diversity index,
are to be encouraged.
153
-------
154
The South Basin of Lake Winnipeg - An Assessment of Pollution
Jo-Anne M. E, Crowe, Canada
The author has presented data in an attempt to demonstrate that Lake Winnipeg is suffering from
an increase in the rate of eutrophication. Three indices were utilized in this endeavor; physical and
chemical measurements, benthic organisms, and fish.
There is little doubt that the lake, as described by the author, is suffering from the effects of
pollution; however, the data presented is not conclusive with respect to changes noted in severely
affected eutrophic lakes. It is unfortunate that no measurements of phosphorous or nitrogen were
available that could be assessed for their role in the apparent changes occurring in the lake.
The author mentions the severe oxygen depletion noted in the Lake Erie basin and the increases in
various ionic species indicative of the conditions in that basin. She then states that dissolved
oxygen values in the lake are usually above 80% of saturation and that the only ionic component
showing an increase was calcium. The Secchi disc readings were significantly reduced in the
39-year period, however.
Examination of the data fails to reveal significant differences in the chemical measurements at first
glance, even of calcium. It would be interesting to subject the data to a statistical analysis in order
to determine the statistical significance of the changes, if any, on a quantitative basis. Mere
comparison of averages means little with this type of data. Differences in sampling techniques
should also be accounted for.
Measurements of the phytoplankton would have been interesting, as would discussion and analysis
of some of the chemical data.
The only really significant changes that were noted were an increase in benthic density, a shift to
more tolerant forms of bottom organisms, and a decrease in the fish catch. Again, a statistical
analysis of these factors would be of considerable value in determining the real significance of
these changes.
Another informative evaluation of the data would be to utilize some form of species diversity
index with respect to the bottom organisms, which will be mentioned later.
While the fishing yield has been significantly reduced, the author states that this reduction could
be attributed to "over-fishing" and the use of illegal methods.
An additional point of interest is the possibility of mercury playing a role in the change in the
benthos. The toxicity of mercury is extreme, and if the concentrations in the fish have "exceeded
0.5 ppm", one cannot help but wonder what the concentrations might be in the organisms being
ingested by the fish.
-------
155
Eutrophication in Some Lakes and Coastal Areas in Finland
With Special Reference to Polyhumic Lakes
Pasi O. Lehmusluoto, Finland
A good review of the phytoptankton production in Finland and in arctic areas was presented by
Dr. Lehmusluoto. He points out that in Alaska, almost half of the annual phytoplankton primary
production may occur beneath the ice sheet in the spring time. He suggests that the mean
maximum primary production rate per unit volume in the growing season measured in situ, or in
constant light, should be used as an index for the many kinds of water bodies to give a relative, but
objective, index for the phototrophic level of the water. It is also demonstrated that the direct
comparison of trophic states of humic and nonhumic lakes is impossible, as the classification of
polyhumic lakes on the basis of phytoplankton production cannot give an objective result.
Polyhumic water in Lake Hakojarvi showed that nitrogen was the primary limiting nutrient and
was also a primary limiting nutrient for bacterial growth. The addition of nitrogen and phosphorus
caused a large algal growth. Seven in situ plastic test cells were used, the cells being ten meters
deep, 1.2 meters in diameter, and having open tops with a water volume of 12 m3. Nutrient
additions ranged from 0.05 to 1.0 mg/l nitrogen and 0.005 to 0.1 mg/l of phosphorus.
Phytoplankton primary production was measured by the carbon 14 method. Phosphorus alone did
not cause significant eutrophication and there were not any large changes when only nitrogen was
added. In the cells where both nitrogen and phosphorus were added, the greater the addition of
nutrients, the larger the increase in the phytoplankton production.
Humus may be important in the total production of polyhumic waters due to light extinction. It is
also proposed that humus, as such, or transformed to bacterial biomass, may serve as food for
zooplankton and, thus lead to relatively high fish production. It should be noted that bacterial
numbers in humic waters are normally comparable to numbers in eutrophic lakes.
In Lake Saimaa, phytoplankton production has been studied. The lake receives 1,200 m3/day of
biologically purified domestic sewage and 260,000 m3/day of mechanically purified pulp mill
waste, which is 44% sulfite and 56% sulphate liquor. The domestic sewage stimulated the algae
growth. Pulp mill effluents were almost lethal to algae at 10% effluent concentrations, and in 1%
and 0.1% effluent concentrations, the algae growth was slightly stimulated. Pulp mill effluents
seemed at first to hinder phytoplankton primary production, but caused pronounced
eutrophication about 9 km from the outfall. Both domestic sewage and pulp mill effluents do
cause eutrophication. The maximum eutrophication caused by the pulp mill effluents is more
intensive than that caused by the domestic sewage, though it occurred far from the discharge area.
Eutrophication is a problem in the coastal area in Finland, as well as in lakes. The major problems
are near the cities on the coast and are important recreational areas. Blue-green algae blooms
occurred during late summer and restricted the use of the waters.
The phytoplankton production 1 km from the outfall was about 30 times higher than that at 12
km distance from the outfall. Annual primary production near the outfall was 5 times higher than
in the unpolluted areas.
In conclusion, it is important to consider all the possible efforts to reduce nutrient input into
receiving waters, in any form, in order to avoid overeutrophication, of waters.
-------
156
It would be interesting to hear the speaker's comments on the recent hypothesis of Knentzel,
which has caused considerable concern in the U. S.
Knentzel (1969) concluded that phosphorus and nitrogen were not the limiting factors in
eutrophication, but instead, organic material and the resulting production of CO2 by bacteria.
Knentzel's conclusions were based on the following observations:
1. In natural water, blue-green algae and bacteria are always found in close association.
Separation is detrimental to algae.
2. Massive algal looms are always associated with excessive amounts of decomposable
organic matter.
3. CO2 is the major nutrient for algal growth. 2 gms CO2 are required for one gram algae.
4. Large amounts of CO2 required cannot come from the atmosphere or land-dissolved
carbonate salts.
5. Bacteria can supply 20 mg/l CO2.
6. P is widespread in nature and algal blooms have been documented where P< 0.01 mg/l;
and no blooms have been documented where P > 0.01 mg/l but no organics.
It is interesting to note that recent work by the Federal Water Quality Administration Laboratory
in Athens, Georgia, appears to lend credence to Knentzel's theory (Kerr, et al., 1970).
Observations on the Recovery Process in a Lake which had Earlier
Received Waste from an Ore-Dressing Plant
Bengt Ahling, Sweden
The objective of this paper was to compare the physico-chemical and biological parameters of
Lake Balsjon in Central Sweden before and after the shutdown of an ore dressing plant. Initially,
the parameters from Lake Balsjon were compared to a control lake before shutdown. The question
asked was "Can a recipient of some waste satisfactorily recover, purify itself to the point at which
its water can be used for more desirable purposes, within a reasonable time?"
The ore dressing plant's principal ores were hematite and magnetite. The process involved gravity
and wet-magnetic concentrations, hence large amounts of iron were deposited in the lake.
Table 1 shows a comparison of the physico-chemical parameters presented in this paper.
The most interesting part of the paper, and possibly the most important factor, is the development
and changing of the ecosystem within the lake. Originally the lake was a 'sterile milieu'. Initially,
there was a noted absence of sessile living lake-bed organisms and submerged plants which was due
to the reduced penetration of sunlight. No zooplankton were found whatsoever.
-------
r*.
tn
TABLE 1
Parameter
Before
Control Lake Lake Bal
After
Control Lake Lake Bal
Comments
1. Turbidity
2. Colour
3. Secchi disk
transparency
4. pH
5. Conductivity
6. Total iron,
mg Fe/1
7. Oxygen
8. Consumption of
permanganate
mgKMn04/l
9. Nitrogen and
phosphorus
<100 ZP units >1000 ZP units 100 ZP units
30 mg Pt/1 650 mg Pt/l 30 mg Pt/1
2-4 meters Few centimeters
7.8 -8.1*
7 times the
control lake
3 meters
<5
20
1-10
90% sat
10
400 ZP units
30 mg Pt/1
2 meters
Reduction in pH (spring)
Increase in pH (fall)
3 times the control lake
,5
Still high
30
30
Returned to
normal hue
Due to reduced
turbidity
Primary production
Increased which
increases pH
Actual values
undecipherable
Bottom sample
showed oxygen
deficit down to
17%
Increase showed
increaise in organic
material in lake
No change
No values reported
in paper
* After dilution with Lake Balsjon waters
-------
158
After the plant shutdown, consequently reducing the turbidity of the water, the lake was again
suitable for biological growth. The occurrence of phytoplantkon was markedly increased,
especially diatoms. Organisms that were completely absent initially began to appear in abundance,
e.g. chrysomonad rhodomonas and dinobryon divergens. Besides increasing the number of
individuals in a species, the number of different species increased. Two years after the plant
shutdown the composition of zooplankton was almost identical to that of the control lake. Also, it
was noted that fauna normally living on the lake bed was in the process of being built back up.
After observation of the physico-chemical parameters, one is led to believe that Lake Balsjon is
returning to its normal state as compared to the control lake. However, the values reported in the
paper are very general in most respects. Yet, the building up of a balanced biosystem is quite
evident by the changing biological picture. Not only is the number of individual species increasing
but the number of species is also increasing. Therefore, one can conclude for this particular lake
that the physico-chemical aspects are returning to a normal pattern slowly, while the ecosystem of
the lake is returning to a balanced system much faster.
As previously mentioned, the use of some form of a "Species Diversity Index" for comparative
purposes would add considerably to the conclusions that could be made from this study.
Depletion of Oxygen by Microorganisms in Alaskan Kivers at Low Temperatures
Ronald C. Gordon, USA
This paper demonstrated that bacterial degradation occurred in subarctic rivers at temperatures of
0° C to 20° C. As the incubation temperature was decreased, a lag phase increased, but extensive
metabolic activity was observed at all temperatures.
The rate also decreased with decreasing temperatures as might be expected. The rate of activity
was affected by the nature of the substrate, primary effluent, and rapid D.O. depletion was shown
at all temperatures, while secondary effluent showed activity at 10° C and 20° C but none at 0° C,
probably indicating the effect of nitrification, i.e., the high temperature dependency for
nitrification yielded no growth at 0° C.
Psycrophillic bacteria grow well at low temperatures, but the rate of activity increases with
increasing temperature.
It is obvious that the 5-day 20° C BOD test is not a good measure for pollution control at low
temperatures because of possible different mechanisms of metabolism. It would appear that the
"yardstick" for pollution in cold regions should be some form of total carbon analysis or at least
that the BOD values measured should be at the temperature of the receiving water.
One wonders why the author did not extend the data analysis to include determination of the
effects of temperature on the rate constants and a comparison of the rate constants themselves. It
would appear that the data required for this information has already been acquired and all that is
needed is some mathematical analysis and then a quantitative presentation of the results.
-------
159
This information would be quite useful inasmuch as the characteristics of the oxidation rate
constants at low temperatures need elucidation.
Prediction of Dissolved Oxygen Levels in the South Saskatchewan River in Winter
Robert C. Landine, Canada
This paper is interesting inasmuch as some of the differences in stream analysis in cold climates
over warm weather conditions are discussed. The low temperatures, ice cover, lack of local inflow,
high oxygen saturation values and low reaction rates are all different from those usually
encountered.
The approach described is more or less a standard one, with modifications for cold weather
treatment. It should first be pointed out that utilization of this type model requires steady state
conditions, an assumption that is rendered invalid because of the hydroelectric power plant
operations.
In addition, the goodness of fit of the proposed model is questionable inasmuch as 4 sampling
points are hardly sufficient for a 202 mile reach. Also, since the change in Dissolved Oxygen was
less than 1 mg/1 in ~12, almost any coefficients would produce a fairly good fit with the relatively
low organic loadings used. Thus, there is no assurance that the model will accurately predict
effects of major changes in flow and/or loading conditions.
The apparent anomaly wherein the assumed values of Kj had a negligible effect on the
deoxygenation rate in the river was partially due to the incorrect assumption that the bottle Kt
was identical to the river deoxygenation coefficient, Kr. This is very seldom true inasmuch as
oxidation in a bottle is hardly similar to that occurring in a river.
The ultimate first-stage BOD is a fixed parameter and should be determined independently by a
long term BOD test. It can be estimated from short term BOD's if the bottle coefficient, KI, is
also determined. L is not a function of Kj as implied by the author. Once L is determined, the rate
of deoxygenation in the river is assumed to be proportional to L and K^, which should be
determined in the river.
It should also be noted that the author's use of the English reaeration equation will probably not
be adequate for the conditions of the Saskatchewan River. This equation was empirically derived
from very small streams and brooks and has been shown many times to be inapplicable to rivers
the size of the Saskatchewan. If the author had to use an equation of this type, he would have
better success (optimistically) using Churchill's equation (1962):
k2 = 5.026 U'
which was derived from what is considered to be the best stream data ever taken. In either case,
the use of empirical equations should be subjected to extreme cases in order to avoid gross errors.
The author might well have applied a more rational equation such as the one progressed by
Thackston & Krenkel (1969):
-------
160
k2 = 0.000125 (UF1/2)U.
h
U
F = 0^ = hSeg
9"
which has a theoretical base and has been successfully used in field measurement.
The comments herein should not detract from the effort put forth by the author inasmuch as
obvious limitations were imposed because of a lack of adequate time and funds. As is the case in
many water quality investigations, the results must be in accord with economic limitations of data
collection.
Pollution - A Biological Study of Some Receiving Waters in Hokkaido
Matsunae Tsuda, Japan
The objective of this paper was to study the pollution effects on the biota of the Ishikari River.
The paper gives a fairly good descriptive and qualitative piece of work but gives no real definitive
idea of the total effect on the biota, the degree of damage, or any way of comparing this data with
other work.
A major weakness in the paper is that the writer has but one source to compare and give
background to the study. A major aspect appears to be concerned with population dynamics. It
should be noted that in 1964, Silvey and Roach published a definitive paper on population
dynamics in fresh water systems using ten years of accumulated data. There has been a great deal
of work dealing with species diversity as related to pollution perturbances. Two fine papers were
published by Odum, et al. (1963), and Pearson (1967), and demonstrate the variances of species
diversity attributed to various pollutants.
The author did mention how and with what equipment the data was taken. Since no reference was
made to sampling techniques and methods of analysis, scientifically, this data has little meaning
either qualitatively or quantitatively. It is obvious that sampling and analysis techniques should be
described so that the degree of accuracy and precision of observed data may be ascertained.
If the microbenthotic organisms were analyzed by a microscope, the species diversity could easily
be obtained. This would yield a somewhat quantitative answer rather than the qualitative answer,
"There were numerous organisms or just a few species were observed". A common species
diversity formula used is:
1 = -Sj Pi Loge Pi
where Pi = n-t/n
n. = number of individuals per species
N = total number of species
-------
161
When working with microorganisms such as zooplankton, etc., Copeland (1966) used a simpler
method to obtain a species diversity, which is:
I = number species/1000 microorganisms
It is also of value to determine some abiotic analyses such as TOC, BOD, COD, and dissolved
oxygen, in order to obtain a better picture of the water quality. This additional information would
allow interpretation of the biological data with greater significance than only qualitatively.
Chemical Effects of Decomposing Salmon Carcasses on Aquatic Ecosystems
David Brickell & John Goering, U.S.A.
The authors have presented some interesting data in an attempt to explain a most complex system.
The toxic nature of ammonium ions in the water is certainly of concern; however, the authors did
not measure pH, which has a rather drastic effect on this toxicity. It is possible that a high
concentration of ammonium ions at a low pH value will be amenable to fish life, but if the pH
value is increased the toxicity will probably increase. It has been reported that the toxicity of a
specified concentration of ammonium compound tested with fishes increased by at least 200%
between pH 7.4 and 8 (1937). It should also be noted that the toxicity of ammonia is reduced in
the presence of CO2 and increased with low concentrations of oxygen.
The authors did not mention the stratification occurring in some of the sampling areas.
Examination of the temperature and the salinity data demonstrates density differentials sufficient
to maintain a rather stable stratified flow regime. This would help to explain the observed
increases in nitrogen and decreases in oxygen that occurred at these lower levels. With the new
emphasis on water pollution control, the nitrogen problem will become more significant as
increased treatment introduces effluents of a more stabilized carbonaceous water and readily
available organisms that can oxidize ammonium to nitric and nitrite to nitrate. It would therefore
be quite useful for the authors to attempt to quantitize the nitrogen phenomena that is the subject
of their paper.
Since it is an autocatalytic reaction, nitrogen transformation can be described by:
dc
d7=KC(N-c)
where: C = concentration of ammonium or nitrate
N = concentration originally present
K = reaction coefficient
Wezernak and Gannon (1968) have utilized this model with success on several receiving waters.
Knowing the time of passage, samples are collected at a minimum of three points in the study
section. One sample is collected at the beginning of the study section to determine the upper limits
of ammonium and nitrite oxidation and to determine the amount of inorganic nitrogen oxidation
between the starting point and the other two stations. Of the remaining two samples, one is taken
-------
162
near the beginning of the study section and one near the end.
The effect of temperature on the process is evidently quite pronounced inasmuch as lower
temperatures appear to significantly decrease the rates of nitrification. It would be interesting to
observe the rate constants for the nitrification processes under the cold conditions of this study.
Optimum temperatures for the nitrifying bacteria have been reported in the range of 28° - 36° C,
considerably above the temperature of this study.
Self-Purification of Polluted Lakes in Temperate Regions - Phosphorus Binding Mechanisms
Jan Werner, Sweden
The author discussed a mechanism for phosphorous removal in lakes. By using solubility products
and thermodynamic relationships, he concluded that ferric ion will be the dominate species of iron
in oxygen-containing waters, and ferrous ion will dominate in oxygen-deficient regions. He also
discussed the pH dependence of ferrous precipitates. It was concluded that phosphorous removal
could be improved by a lowering of the pH to 4 because of the increase in stability of the binding
of phosphate to iron. These conclusions can also be found in work by Morgan and Stumm (1964),
Hem and Cropper (1959), Anon. 0970), and others.
It should be noted that a substantial fraction of iron in lake waters is present in suspended form
and the insoluble ferric ion sediments into the hypolimnetic waters. The rate of sedimentation is
influenced by many factors, probably the most significant being the colloidal chemical nature of
the ferric precipitate.
The ferrous ion may be soluble up to a few mg/l, depending on alkalinity and pH and is of
significance in the oxygen-deficient hypolimnion. The ferric ion not reached in the overlying
waters will be reduced at the mud-water interface.
The progressive accumulation of phosphate in the hypolimnion may be partly attributed to
plankton, but other factors are involved. There is usually more iron than phosphorous in lake
water, leading to the formation of ferric-hydroxy-phosphate in the upper waters immediately after
circulation. Hutchinson (1947) has reported that an oxidized mud surface not only holds
phosphate but prevents diffusion of phosphate from deeper mud layers, as ferrous iron is always in
excess and when oxidized, precipitates the phosphate.
The author did not mention the amphoteric properties of ferric hydroxide, which may play a role
in the process: Isoelectric point at a pH = 5.5, pH > 5.5 the species is more negative, pH < 5.5 the
species is more positive. Also, the various exchange mechanisms in the sediment and on the
hydrous metal oxides should be investigated. Since silicates interact with iron in a manner similar
to phosphates, their role should be elucidated.
The author also failed to take into account the species change of the inorganic phosphate with
changes in pH. At very low pH's, H3PO4 is the major species, at approximately a pH of 4, H2PO4"
is the major species, and so on until at pH's of 11 to 13, orthophosphate becomes the predominate
inorganic phosphate species. At a pH of 4, the ferric cation will not precipitate as FeP04 but as
Fe(H2PO4}3, and this precipitate will have a completely different precipitation solubility product
-------
163
tfian ferric phosphate. In a chemical sense, we are dealing with a dynamic and unsteady system,
and, therefore, the rates of precipitation equilibrium studies carried out in this paper.
Finally, the influence of biological activities on the chemical reactions described should be
investigated. There is little doubt that the biota plays a significant role in determining the
distribution of chemical constituents in a lake.
REFERENCES
Anon. (1970) Chemistry of nitrogen and phosphorus in water, J. Amer. Water Works Assoc.
Churchill, M. A. (1962) The prediction of stream reaeration rates, Tennessee Valley Authority,
Cattanooga, Tennessee.
Copeland, B. J. (1966) Effects of industrial waste on the marine environment, J. Water Poll.
Control Fed.. 38, p 1000.
Ellis, M. M. (1937) U. S. Department of Commerce, Bureau of Fisheries, Bull. 22.
Hem, J. D. and Cropper, W. H. (1959) Survey of ferrous-ferric chemical equilibria and redox
potentials, U. S. Geol. Sur. Water Supply Paper 1459A.
Hutchinson, G. E. (1957) A treatise on limnology. Volume I, Wiley and Sons, New York.
Kerr, P. C., et al. (1970) The interrelation of carbon and phosphorus in regulating heterotrophic
and autotrophic populations in aquatic ecosystems, FWQA, Southeast Water Lab, Athens,
Georgia.
Knentzel, L. E. (1969) Bacteria, carbon dioxide, and algal blooms, J. Water Poll. Control Fed.,
Vol. 41, No. 10.
Morgan, J. J. and Stumm, W. (1964) The role of multivalent metal oxides in limnological
transformation, as exemplified by iron and manganese, Proc. Second Inter. Conf. on Water
Poll. Res., Pergamon Press.
Odum, H. T., et al. (1963) Diurnal metabolism, total phosphorus, ohle anomaly and zooplankton
diversity of abnormal marine ecosystems of Texas, Inst. Texas Mar. Sci.
Pearson, E. A., et al. (1967) Pollution and marine ecology, Interscience, New York.
Slvey, J. K. G. and Roach, A. W. (1964) Studies on microbiotic cycles in surface waters, J. Amer.
Water Works Assoc., p 60.
Thackston, E. L. and Krenkel, P. A. (1969) Reaeration prediction in natural streams, Proc. Amer.
Soc. Civil Engineers, Vol. 95, No. SA1.
Wezernak, C. T. and Gannon, J. J. (1968) Evaluation of nitrification in streams, Proc. Amer. Soc.
Civil Engineers, Sanitary Engineering Div.
-------
THE INFLUENCE OF TEMPERATURE ON THE REACTIONS
OF THE ACTIVATED SLUDGE PROCESS
Pal Benedek and Peter Farfcas
The rate of biochemical reactions is increased by an increase in temperature and vice versa.
However, while the microbial processes of fermentation industries are carried out at constant
temperature, biological waste treatment is exposed to temperature changes predetermined by the
prevailing climate. This fact, of course, exerts an influence on purification efficiency. Namely,
detention times in the waste treatment facilities are fluctuating in summer and winter between the
same limits, while the rate of the reaction - and so the purification efficiency - decreases
considerably with falling temperature.
Temperature dependence of the biological treatment process can be described with a simple
mathematical model only between certain limits because it is a complex phenomenon. It is well
known that various technological modifications of the activated sludge process react diversely to
changes in temperature, the aerated lagoons being most sensitive to the cold (Table 1).
Explanation of this phenomenon was given by Eckenfelder (1966) who supposed that the rising
temperature results in anaerobic conditions in the inner part of the floes (oxidation lagoons have
dispersed bacterial systems, whereas activated sludge plants have flocculated ones). Thus, reaction
rate increase with temperature is compensated for in a negative sense by the smaller bacterial mass
taking part in aerobic substrate removal processes.
The situation is further obscured by observations of Wuhrmann (1964), indicating that not only
the rate of the reaction, but also its stoichiometric constants (g O2 consumed/g substrate removed)
are changing with temperature. Downing (1968) pointed out that this was caused predominantly
by the intensive temperature dependence of nitrification. (At temperatures above 10° C nitrogen
metabolism produces NO3, whereas at lower temperatures NH^ is formed, the latter process
needing less oxygen.) It cannot be excluded, however, that the fraction of substrate, being
oxidized for energy yield, is also changing with the temperature.
The aims of this study are to reveal the physico-chemical basis of temperature dependence
measurements, and based on the kinetic interpretation, to measure the temperature dependence of
individual substrate removal and oxidation processes.
KINETIC INTERPRETATION
In a non-steady state system, the reaction rate,-^ is a function of temperature, T° (in degrees
Kelvin), and reactant concentration. St, at time t, as described below:
ds n
— = KS" (1)
where^is f(T°, St) and K is f(T°); *
where K is the reaction constant (being a function of temperature) and n is the order of reaction.
164
-------
165
TABLE 1
Temperature Dependence of Biological Waste Treatment Processes
Temperature
Process range, C° Q10 9 Source
Activated sludge 0 - 20 1.00 -1.48 1.000 -1.040 Eckenfelder, 1966
THckling filter 0-20 1.41 1.035
Aerated stabilization pond 0-20 1.96-2.15 1.07-1.08
Piper and pulp mill wastes 2 - 30 1.35 -1.56 1.031 -1.046 Carpenter et al., 1968
To describe temperature dependence of the rate of enzyme-catalyzed biochemical reactions -
where one cannot speak of an "order of reaction" in the strict physico-chemical sense - Equation 1
modifies to:
dS
-ft = A = K • X ' f (St) (2)
where both ^ and A are f(T°, St) and K is f (T°);
where x is the concentration of biocatalyst (enzyme or active bacteria) and f(St) refers to the
substrate dependence of reaction rate, that is, "biological activity", A.
Based on the foregoing generalized relationships, the Michaetis-Menten model (Pearson, 1968;
Benedek and Parkas, 1968) can be written:
x-^
where K = Vcmax as f(T°); *
AH
and
so K = Vmax = (constant) e RT (4)
§t
f(SJ = - (5)
* S. + Km
Equation 4 represents the Arrhenius equation, where AH is the "activation energy" expressed in
calories (Thimann, 1964; Johnson, et al., 1957; Ingraham, 1962). Vmax is the maximal value of
specific substrate removal rate defined as-rr" ^j|. Equation 5 is the so called "enzyme kinetic
function", where Km is the Michaelis constant.
Supposing that St < Km (which is the case under normal operating conditions in the activated
sludge process), the generally known first-order reaction rate can be attained:
yrnax
A =— *X'St = K'-X'St (6)
Km
-------
166
where K' is a first-order reaction constant and n = 1.
From the point of view of temperature dependence measurement, it is important that x and f(St)
should be constant. In this case, from Equation 2 it follows that
£l = Jii
A2 K2
where the subscripts 1 and 2 refer to A and respective K values measured at two different
temperatures. Metabolic rate measurements must be made at different temperatures with the
substrate being removed on the course of a zero order reaction - in this case, reaction rate does not
depend on substrate concentration, or St should be the same constant value for all measurements
performed. Of course, sludge samples have to be taken from the same system, thus assuring
constant bacterial concentration (x), too. Adopting the above test conditions, the actual biological
activity (A), can be measured instead of the respective K values, which are difficult to develop.
Metabolic reaction rate parameters like substrate and endogenous respiration, substrate removal,
dehydrogenation, etc., are regarded as different forms of A.
To calculate AH from Equation 4 and 7, it follows:
A2 T2 • T,
AH = log • 2.3 R • (8)
A, T2-T,
As an index for temperature dependence, the coefficient 6 is also used with the following
definition of Eckenfelder (1966):
— (T T )
A! ~ *9'
A further symbol characterizing temperature dependence is QIC, the factor indicating how many
times the reaction will increase if the temperature is raised by 10° C (Thimann, 1964).
A(T° + 10)
Q,o = (10)
An interrelation between the three parameters is given below, and deducted in the Appendix:
logQ10 = 10log0 = AH (2.54) (10'5) (11)
For steady state systems, like continuously working activated sludge plants, hydraulic residence
time V/q must be employed instead of physical time, t, and S , effluent substrate concentration,
instead of St-
The term describing removal activity follows from the materials balance. Equation 2:
-------
167
where V is the aeration compartment volume, q is the waste flow, and S is the substrate
concentration in the raw waste. The same rationale is employed as in Equations 2-6, observing the
above indicated changes in t and St, respectively. Assuming a first order reaction analogous to
Equation 6. the expression can be written as:
where K' = f (T°),
from which the value of K' can be calculated graphically after rearrangement:
So V
= 1 + K'-X (14)
Se q
and plotting the respective K' values versus T°, using either the rearranged Equation 8 or the
logarithmical form of Equation 9, the temperature dependence may be calculated (Carpenter et
al., 1968).
Strictly speaking, in Equations 12-14, the term Se - Smm, representing available substrate
concentration, should be taken for Se, where smm is the so-called "residual BOD" remaining in
the effluent after prolonged biological treatment (Benedek and Farkas, 1968 and Benedek et at.,
1968). If the "residual BOD" is neglected, Sfi will exert less change in the temperature function
ttian the "available food concentration", the latter having the proper direct relationship for
removal rate.
EXPERIMENTAL
Measurements were performed on endogenous sludge that had been aerated for some days
previously. Sludge samples were taken from plants treating municipal and pharmaceutical wastes,
respectively. Some tests were also made with sludges grown on pig farm wastes or phenolic wastes.
The temperature range was 0° - 25° C. For a set of measurements, one-liter sludge samples were
taken from the same system, chilled to the desired temperature and thermostatized during the
respiration or activity (substrate removal) measurements. All measurements were done in
non-steady state (batch fed) systems. Temperature dependence of the following processes were
examined:
Colloid removal rate,
Oxidation rate of adsorbed colloids,
Oxidation rate (substrate respiration) of dissolved substrate,
Endogeneous oxidation (respiration) rate.
In the case of dissolved substrate, the amount of O2 consumed per gram of substrate added was
also determined using the short term oxygen demand (STOD) process of Vernimmen et al. (1967).
-------
168
Colloid removal rate was measured mixing a stock colloid substrate solution (caseine, soap or
starch) with endogenous sludge of about 3 g/l MLSS. It was settled for about 5 minutes, then the
supernatant was filtered and St measured as COD. The SQ concentration for t = 0 was calculated
from the well-known mixing formula.
Removal activity. A, was calculated as follows:
A!
A = = (15)
AS SQ-St
The further fate of adsorbed substrate was followed by respirometry. Because the biooxidation
rate of starch was found to be very low (substrate respiration could be neglected in comparison to
endogenous respiration), this test was carried out using a washed caseine suspension. The mixed
liquor containing the caseine was aerated for an hour after mixing to allow the dissolved fractions
to oxidize, which was indicated by an intensive drop in respiration. Respiration rate measured
after the removal of the dissolved fraction was regarded as the net sum of the substrate respiration
due to the biological oxidation of caseine and the endogenous respiration.
Substrate removal rate was measured by means of the Farkas (1968) activity measurement
procedure. Most substrates (acetate, sucrose) were removed by a zero order reaction, thus assuring
f(SJ to be constant. Substrate load was kept constant for each set of measurements.
Respiration measurements were done with membrane-coated D.O. probes, measuring first the
endogenous respiration at the desired temperature, then adding a constant amount of the substrate
to be examined and measuring the total respiration. Substrate respiration was calculated as the
difference between total and endogenous respiration.
The log reaction rates (A) were plotted against 1/T°, (see Equation 6), where reactions obeying the
Arrhenius law give straight lines. In all cases where this did not occur, the QIO and 0 values were
calculated for the 0° C - 10° C and 10° C - 20° C temperature ranges, using Equations 10 and 11.
RESULTS
The rate of starch removal - which was a purely adsorptive process - did not show any significant
thermal dependence. From Figure 1 the AH value was calculated and found to be 364 calories,
which is much less than AH values encountered in biochemical substrate removal processes.
The plot of the biochemical oxidation of adsorbed caseine suspension (Figure 2) indicates that
biooxidation of caseine obeys the Arrhenius law.
The metabolic rate of dissolved substrates did not give straight lines in the Arrhenius system, as
can be seen in Figure 3 in the case of activated sludge grown on municipal waste and fed acetate.
Nearly the same result was obtained with the same sludge and sucrose (Fig. 4) and activated sludge
grown on a pharmaceutical waste, fed phenol and acetate, respectively (Fig. 5).
In contrast to the above, endogenous oxidation always followed the Arrhenius law. The thermal
dependence of endogenous respiration of four different kinds of activated sludge is plotted in
-------
169
3,327
JBSO-
25 2O
s
—I—
3*0
450
460
Reciprocal of ota. iarr^perolurs
FIGURE 1 Temperature dependence of starch removal rate
-------
170
Terryjerafun? [C.']
0 S
Reciprocal of ota temperokjre
endogenous 7/ „>]
65O rng/i cneese suspension I < J
32SO mg/i cheese suspension
FIGURE 2 Temperature dependence of the biological oxidation rate of cheese suspension
Figure 6. Respiration at 20° C was arbitrarily taken as unity to make comparison easier. The
activation energies were within the range of AH = 13,150 ± 550 cal.
All thermal dependence results obtained in this study together with some representative data from
the literature are included in Table 2.
In Figure 7, the stoichiometric ratios of substrate oxidation, (g 02/g substrate) obtained with
STOD measurements are plotted versus temperature.
For acetate, this ratio seems to remain unaffected by temperature. The value of 0.4 g 02/g acetate
is in good agreement with the results of Warburg's measurements and data of other authors. For
phenol, the stoichiometric ratio ranged from 0.7 to 0.9, the smaller figure being valid at 0° C.
-------
171
437
f&crian:
*, Ace/ate load
2, Activated sludge grown on municipal waste
FIGURE 3 Temperature dependence of the biological oxidation rate of acetate
-------
172
500
Suu-c&e load CO mg/i
2, Activated ilrtjp (f-cntn an
FIGURE 4 Temperature dependence of the biological oxidation rate of sucrose
-------
173
fSedpracal of ate. ierrperature
FIGURE 5 Temperature dependence of the biological oxidation rate of phenol and acetate,
respectively
-------
174
& H• OtOO±S5Ocol
Qjgin of activated jMjQQe
e Liquid -sirrte ctng
Q FtarmcLcenticat +phencto: tmstf
+- MurKJpot waste
$ Endajencui
2,MLSS 2-5 gft
t5-2O
FIGURE 6 Temperature dependence of the endogenous respiration of four different activated
sludge samples
-------
175
DISCUSSION
The thermal dependence data were in good agreement with those of other authors. Johnson et al.
(1957) statistically analyzed the available body of experimental data comprising a broad variety of
biological processes. They found that two distinct peaks exist in the frequency of occurrence of
AH values: one between 11,000 - 13,000 cal, (from the results reported herein, the AH values of
dissolved substrates above 10° C - and perhaps, endogenous respiration - belong in this group), the
other between 15,000 - 18,000 cal, where the removal of dissolved substrate belongs, at
temperatures lower than 5° C.
According to the Crozier theory, substrate metabolism cannot be described with one simple
Arrhenius equation, if it is assumed to be a complex process comprising a series of coupled
biochemical reactions. At different temperatures there would be different "master reactions"
characterized by different activation energies. (The reaction determining the overall reaction rate is
called "master reaction" by Johnson et al., 1957). The temperature value, where there is a break
TABLE 2
Temperature Dependence of The Rate of Metabolism For Activated Sludge
Metabolism
1. Adsorptive removal rate of
starch
2. Biological oxidation of caseine
suspension
3. Removal rate of acetate
4. Substrate respiration with
acetate
5. Removal rate of sucrose
6. Substrate respiration with
sucrose
7. Removal rate of phenol
8. Substrate respiration with
phenol
9. Endogenous respiration
(4 different sludges)
10. Respiration of activated sludge
11. Rate of nitrification
12. O2 • consumption rate when
determining BOD
Temperature
range, C°
0 -25
0-25
AH
cal
364
Q10 . 6
1.022 1.000
Source
Fig. 1
17.000 2.66 1.103
Fig. 2
0
10
0
10
0
10
0
10
0
10
0
10
0
0
-10
-25
-10
-25
-10
-20
-10
-20
-10
-20
-10
-20
-
-20
-25
.
7.200
33.700
7.200
37.000
.
-
_
-
.
-
.
1
.51
6.90
1
8
5
2
6
2
3
1
3
.51
.42
.80
.86
.50
.72
.87
.72
.18
2.0
13.150
-
2
2
2
3
.15
.25
.04
.81
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.042
.214
.042
.242
.192
.110
.206
.105
.131
.056
.134
.072
.078
.085
.074
.143
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
3
3
3
3
4
4
4
4
5
5
5
5
6
Eckenfelder
Wuhrmann,
,1966
1964
Downing, 1968
4 -20
20 -30
3.55 1.135 Eckenfelder, 1966
1.72 1.056 Eckenfelder, 1966
-------
176
• ptenoi
x acetate
Ifl
OS
x
«
•
•
1 iamb, et at'.[e]
Onto
FIGURE 7 Temperature dependence of substrate oxidation oxygen demand
on the Arrhenius plot, is called the "critical temperature", where two master reactions are
supposed to "switch over." In Figure 3, a sharp break at 7° C can be observed on the Arrhenius
plots of both acetate removal and substrate respiration, and with all dissolved substrates, a more or
less pronounced "bend" can be seen in the range of 5° C - 10° C.
It is known that dehydrogenases have AH values of 1 1,200 calories, and cytochromes (respiration
enzymes underlying cyanide poisoning) have AH = 16,000, but the data is too fragmentary to
conclude that above 10° C the master reaction is dehydrogenation, and at 0° C,
cytochrome-catalized respiration.
The Arrhenius plot of the metabolic reactions of activated sludge is similar to those of
psychrophylic bacteria that exert a technologically utilizable metabolic activity even at 0° C
(Ingraham, 1962).
Strikingly, the biooxidation of adsorbed food readily follows the Arrhenius law. According to
present knowledge, adsorbed food must be hydrolyzed and transformed into dissolved compounds
prior to biological utilization. Considering this, it remains an open question why caseine oxidation
shows a temperature dependence different from the dissolved substrates.
Endogenous respiration also does not show any "critical temperature value." It seems plausible
that endogenous metabolism has only one master reaction in the 0° C * 25° C temperature range.
Endogenous "basic" metabolism has to be very stable because it is more important from the point
of view of survival and resistance against adverse environmental conditions than substrate
metabolism. It may be assumed that endogenous metabolism, as a well-stabilized biochemical
reaction chain, has only one "master reaction", consequently behaving as one simple reaction.
-------
177
The amount of substrate oxygen demand changes little with temperature. This is in fair agreement
with the findings of others (Vernimmen et al., 1967 and Lamb et al., 1964} (Fig. 7). This means
that in the instances studied only the reaction rate, but not its stoichiometric constants, has
changed with the temperature. Similar conclusions were drawn from BOD curves recorded at
different temperatures (Gotaas, 1948; Wilderer et al., 1969; Chia, 1969).
Final conclusions concerning thermal dependence of the activated sludge process can be drawn
only from the continuous process. The calculation of K;, according to Equation 12, is possible
only if a large collection of data is at hand. The S /Se ratio usually has a great fluctuation, and the
data must be treated statistically (Wuhrmann, 1964).
CONCLUSIONS
1. Endogenous metabolism follows the Arrhenius law, while the metabolism of dissolved
substrate does not. An explanation to this is sought in the Crozier theory, supposing that at
different temperatures, different master reactions (having different activation energies) are the
"bottlenecks" of the biological reactions.
2. The adsorptive mechanism of activated sludge also has a temperature dependence, but
this amounts to less than 10% of the thermal dependence of biological processes.
3. The value of grams-oxygen consumed per gram substrate oxidized was found practically
constant in the range of 0° - 25° C.
4. Temperature dependence of continuously working activated sludge plants is usually less
than would be expected from non-steady state measurements. From the possible explanations of
this phenomenon, the Eckenfelder theory of partial anaerobiosis within the floes was mentioned.
Further, it should be kept in mind that most wastes contain colloids, the removal rate of which
does not depend on temperature. A great difference may be encountered in the thermal
dependence of treatment efficiency, depending on whether the waste contains colloids or dissolved
compounds, in favor of the former.
5. A great importance has to be attributed to heat storage in water bodies underlying
biological treatment; a system with long detention time may not be economic, because the thermal
reserve - consequently, higher removal rate - of a short detention time system may compensate for
the long detention time with heat loss in a total oxidation plant.
APPENDIX
Interdependence Between The Thermal Coefficients of AH, 0, and Q, 0
From the Arrhenius equation. Equation 4, it can be written:
A2 AH T2 - T,
log— = — ' (A-1)
A, 2.3 R T2 ' T,
-------
178
From the above equation, R = 1.99 Kcal/mol.K°, and T2 ' Tt has the approximative value of
85,850 in the temperature range of 0° C - 30° C. Thus it can be written:
A2 AH T2-Ti AH
log = ' = -(T2-T!) (A-2)
A! (2.3)(1.99) 85,850 394,910
Introducing the 6 constant:
AH
logfl = = AH <2.54)(10'6) (A-3>
394,910
From A-2 and A-3 we obtain:
A2
log = log0(T2 -T,)
or
A2
= 0
-------
179
Chi a Shun Shih and Stack, V. T., Jr. (1969) Temperature Effects in Energy Oxygen Requirements
in Biological Oxidation, J. Wat. Pollut. Control Fed., 41, R461.
Downing, A. (1968) Factors to be Considered in the Design of Activated Sludge Plants, Advances
in Water Quality Improvement, Univ. of Texas Press, Austin.
Eckenfelder, W. W., Jr. (1966) Industrial Water Pollution Control, McGraw Hill Book Co., New
York.
Farkas, P. (1968) Method for Measuring Aerobic Decomposition Activity of Activated Sludge in
an Open System, Fourth Conf on Water Pollution Research, Pergamon Press, London (In
press).
Gotaas, H. B. (1948) Effects of Temperature on Biochemical Oxidation of Sewage, Sew. Wks.
Jour.. 20, 441.
Ingraham, J. (1962) Temperature Relationships, The Bacteria, IV, Academic Press, New York.
Johnson, F. H., Eyring, H. and Polissar, M. J. (1957) The Kinetic Basis of Molecular Biology, John
Wiley and Sons, New York.
Lamb, J. E., Westgarth, W. C., Rogers, J. C. and Vernimmen, A. P. (1964) A Technique for
Evaluating the Biological Treatability of Industrial Wastes, J. Wat. Pollut. Control Fed., 36,
1263.
Pearson, E. A. (1968) Kinetics of Biological Treatment, Advances in Water Quality Improvement,
Univ. of Texas Press, Austin.
Thimann, K. W. (1964) Das Leben der Bakterien Wachstum, Stoffwechsel und
Verwandtschaftsbeziehungen, Fischer, Jena.
Vernimmen, A. P., Henken, E. R. and Lamb, J. C. (1967) A Short-Term Biochemical Oxygen
Demand Test, J. Wat. Pollut. Control Fed., 39, 1006.
Wilderer, P., Hartmann, L. and Janeckova, J. (1969) Der Einfluss der Temperatur auf dem
BSD-Endwert, Vortrag., Universitat, Karlsruhe.
Wuhrmann, K. (1964) Hauptwirkungen und Wechselwirkungen einiger Betriebsparameter im
Belebtschlammsystem. Ergebnisse mehrjahriger Grossversuche EAWAG, Vertrag, Zurich.
-------
TEMPERATURE EFFECTS ON BIOLOGICAL
WASTE TREATMENT PROCESSES
W. Wesley Eckenfelder. Jr. and A. J. Englande
INTRODUCTION
Variations in liquid temperature affect biological waste treatment processes in two interdependent
ways. First, the oxygen utilization rate which reflects the energy transfer of the process has a
defined temperature relationship. This reaction-temperature dependency is reflected by a
temperature coefficient 0. Secondly, in any biological system a relationship will exist between the
oxygen transfer rate to the biomass and the oxygen utilization rate of the biomass. It is therefore
necessary that 6 also consider the effect of temperature on the oxygen transfer rate. The absolute
temperature effect on the performance of a particular process will therefore be the resultant of
these two effects.
In order to estimate the thermal influence on a process, a modification of the Van't
Hoff-Arrhenius equation is usually used:
K= K o fi (T-20)
T 20 C
It should be noted that this relationship is valid only within specific limits. A lower limit is
imposed by retardation of bacterial activity for mesophilic organisms as the temperature
approaches freezing. Relatively high reaction rates may still exist at very low temperatures for
psychrophilic organisms The rate of the biological reaction rates will increase in accordance with
Equation 1 with temperature to an optimum value for most aerobic systems. Further increases in
temperature result in a decreased rate for mesophilic organisms. Maximum biodegradation by
thermophilic organisms, however, will be obtained over a temperature range of 35° C to 65° C.
The effect of temperature on biological processes is the resultant of several factors which consider
the type and distribution of microorganisms, the type of process and the design and operation of
the system.
Bacteria can be broadly classified into three categories as to their response to temperature, namely
psychrophilic, mesaphtlic and thermophilic. Each of these categories have an effective temperature
range. By far, the majority of wastewater treatment processes function in the mesophilic range,
although it has been demonstrated that some processes in northern climates function in the
psychrophilic range. A few systems have been designed and operated in the thermophilic range.
Probably, the most responsive factor relating temperature effect to process performance is the
food/microorganism ratio (F/M). At high F/M values the biomass may be filamentous or dispersed
and 6 will be high indicating a direct temperature effect on each organism. By contrast, at lower
F/M ratios the biomass is flocculated and diffusional mechanisms into the floe become significant.
More of the floe is aerobic at lower temperatures so that a greater aerobic biomass at lower
temperatures is capable of stabilizing almost the same quantity of organic matter as a smaller more
active biomass at higher temperatures. It is because of this that the activated sludge process at
conventional loadings (F/M 0.6) and trickling filters yield a much lower temperature coefficient, 8,
180
-------
181
than those processes using primarily dispersed growths such as the aerated lagoon. It is probable
that the mixing intensity in the process which affects floe dispersion also influences the coefficient
0. Temperature effect on the various processes are shown in Figure 1.
In order to adequately predict the magnitude of the temperature influence on a system's
efficiency, an appropriate 8 value must be employed. The temperature coefficient will vary
depending primarily on the nature of the process under consideration. This paper discusses 6 for
various biological process relationships. The processes considered include waste stabilization
ponds, aerated lagoons, and activated sludge.
I.O
AERATED LAGOONS AND
STABILIZATION BASINS
EXTENDED
AERATION
ACTIVATED SLUDGE
DISPERSED
FLOC
FLOCCULATION
FILAMENTOUS AND DISPERSED GROWTH
RANGE RELATED TO
MIXING INTENSITY
F/M
FIGURE 1 Relationship between temperature coefficient, 8, and food to microorganism ratio,
F/M
DETERMINATION OF 0
The temperature influence on a waste treatment process is usually reported as a change in the
percent removal of BOD. The temperature coefficient 8 which reflects the change in reaction rate
can be computed by graphical methods. Alternative calculations may be employed depending on
the type of data available and the type of process used.
Figure 2 illustrates one method of analysis which is developed as follows for completely mixed
aeration systems:
The BOD removal relationship can be defined as
-------
182
VSe
(2)
in which
S and S are the influent and effluent BOD concentrations respectively
O c
X is the aeration volatile solids concentration
t is the aeration time
k is the mean reaction rate coefficient
and
VSe
= % removal (R)
(3)
Combining Equations 2 and 3 one obtains:
=
1 -R
(4)
2.0
1.0
>• .8
.6
I
_L
10 20 30 40
TEMPERATURE (°C)
50
60
FIGURE 2 Temperature function for activated sludge (Hunter et al., 1967)
-------
183
Let the left side of Equation 4 equal Y. Then, if Xv and t remain constant, kj is proportional to
T Ort
One can rewrite the equation k = ® m tne f°rm
Y = G ' 01"-20 (5)
where G is a constant;
thus
log Y = log G + (T-20) log 0 (6)
As shown in Figure 2, 6 is computed from the slope of a semilog plot of Y versus temperature for
a constant t and Xy.
The temperature coefficient for aerated lagoons where the detention time varies can be computed
from Equation 1:
logkT = Iogk20 + (T-20)log0 (1a)
A semilog plot of percent BOD remaining versus detention time will yield a slope representative of
the reaction rate k. In accordance with Equation 1a, a plot of k versus temperature (as shown in
Figure 4) will yield the coefficient &.
Values for the reaction rates may also be obtained by direct calculation using Equation 2, and the
coefficient & computed.
WASTE STABILIZATION PONDS
Temperature has a significant influence on waste stabilization pond efficiency. Both
photosynthetic oxygen production and biological degradation rates are greatly affected by
temperature variations. While optimum photosynthetic activity is maintained at approximately
20° C, the upper and lower limits appear to be about 35° C and 3° C respectively.
A value for 8 of 1.072 was obtained by Herman and Gloyna (1958) for a temperature range
between 3° C and 35° C. Suwannakarn and Gloyna (1964) found 9 equal to 1.085 for a synthetic
soluble sewage between 9° C and 35° C. Oswald (1966) found a temperature coefficient of 1.075
for domestic sewage between temperature limits of 10° C and 18° C. All the reported values are in
dose agreement, providing an average 8 value of 1.077.
ACTIVATED SLUDGE
Table 1 summarizes 8 values computed from data available in the literature for activated sludge
processes. Temperature coefficients range from 1.0 to 1.041. The value of 1.076 computed from
Vfahrmann (1966) is the probable result of a dispersed growth due to the very high
]food-rnicroorganism ratio of 2.03. Data reported by Hunter et al. (1967) are shown in Figure 2. A
value of 6 was determined to be 1.035 for the synthetic sewage over a temperature range of 4° C
-------
TABLE 1
Temperature Coefficient Evaluation
Activated Sludge
Temp.
Range
C
4-45
5-30
26-37
9-17
10-25
F/M**
0.58
0.44
0.81
0.45
0.57
2.03
0.43
0.22
0.74
MLVSS
(mg/1)
1,600*
1,100
1,100
1,870
3,200*
480*
2,640*
4,800*
800*
A,
623
435
435
750
229
108
124
115
248
Substrate
Synthetic Sewage
(dry dog food meal)
Slurried dog
food meal
Dog food meal
witn dextrose
& gelatin
Phenol
Kraft black
liquor
Domestic Sewage
Domestic Sewage
Batch or Detention
Continuous Time (hr)
Batch
(bench scale)
Continuous
(bench scale)
Continuous
(bench scale)
Continuous
(pilot plant)
Batch
(bench scale)
16
12
3
2.67
10
6 Source
1.035 Hunter et al. (1967)
1.037 Ludzack et al. (1961)
1.041
1.016
1.006 Carpenter et al. (1968)
1.076 Wuhrmann (1966)
1.0
1.0
1.015 Sawyer (1940)
-------
185
to 45° C. Further increases in temperature resulted in a sharp decrease in process efficiency. This
data could be interpreted to show a detrimental effect of temperature below 20° C and above 45°
C. A lower optimal temperature was found by Carpenter et al. (1968) as 37° C. An activated
sludge plant at a West Virginia Pulp and Paper Company plant showed a decrease in process
efficiency at temperatures in excess of 40° C. The activated sludge process appeared to function
satisfactorily at temperatures as low as 4° C.
TABLE 2
Temperature Coefficient Evaluation
Aerated Lagoons
Temp.
Binge MLVSS
°C F/M (mg/1)
10-30 -
13-20 2.51 106*
2-30 2.50 80*
S
(mgTl) Substrate
Cotton Textile
Waste
266 Domestic Sewage
200 Kraft
Neutral sulfite
semichemical
Acid sulfite
Roofing felt
Board mill
Avg. of all 5
<10°C
>10°C
Batch or
Continuous
Continuous
(pilot plant)
Continuous
Continuous
for 2.5 & 5
day detention
Batch
(bench scale)
Detention
Time (days) 6
2.6-5 1.035
8.6 1.046
2.5-10 1.031
1.046
10 1.039
1.040
1.031
1.035
1.058
1.026
Reference
Sawyer (1940)
Eckenfelder (1971)
Carpenter et al. (1968)
* Assumed MLSS = 50% influent BOD
Assumed MLVSS = 80% MLSS
AERATED LAGOONS
Pertinent descrtptives and computed temperature coefficients for the aerated lagoon process are
summarized in Table 2. A range of 9 from 1.026 through 1.058 was observed.
Hie variation in BOD removal characteristics over a temperature range of 10° C - 30° C from
continuous treatment studies by Sawyer (1966) is shown in Figure 3. The temperature coefficient
from these data is 1.035.
-------
186
345
DETENTION, days
FIGURE 3 Temperature function for aerated lagoon (Sawyer, 1966)
Carpenter et al. (1968) conducted an extensive study of pulp and paper mill wastes and calculated
an average 0 of 1.035 for five different wastes over a temperature range of 2° C - 30° C. Figure 4
shows the graphical solution for 9. It is significant to note that two temperature coefficients might
better describe the presented data than the one reported. The same trend was observed for each
waste indicating a 0 equal to 1.026 and 1.058 for temperature limits of 10° C - 30° C and 2° C -
10° C respectively.
Sawyer (1966) and Carpenter et al. (1968) conclude that aerated lagoons with short detention
times will be extremely sensitive to temperature change. This effect is dampened by retention
periods in excess of five days.
There have been recent reports of a relatively small temperature effect on aerated lagoons treating
domestic sewage in northern climates. Two factors must be considered in interpreting such data.
Rrst, the major portion of the BOO in domestic sewage is present in suspended and colloidal form
and a primary removal mechanism is adsorption and flocculation. These mechanisms are relatively
insensitive to change in temperature. Secondly, the retention periods in most aerated sewage
lagoons are long and changes in removal due to temperature are wasted. It is the writer's
contention that further studies on temperature effects of these systems are needed before valid
conclusions can be drawn.
-------
187
0.7
0.6 -
0.5
0.4
>
o>
0.3
0.2
6=1.026
REPORTED VALUE
PROPOSED VALUES
J_
20 3O
TEMPERATURE (°C)
0 10
FIGURE 4 Derivation of temperature coefficient for aerated lagoon
4O
-------
188
DISCUSSION
The summary presented in Table 3 indicates the appropriate 0 and corresponding temperature
range for the various aerobic biological processes employed in wastewater treatment. Variations in
the temperature coefficient can be rationalized by considering the inherent difference in process
operation.
The parameter most influential in determining process temperature sensitivity is the food to
microorganism ratio (F/M). The characteristics of the biomass will be dictated in major part by
F/M. At very high values of F/M the biomass is dispersed. At intermediate organic loading levels
dispersed filamentous growths may develop. At low loading levels, flocculation occurs, resulting in
the formation of gelatinous floes containing millions of individual organisms. Under starvation
conditions floe dispersion will result. In the activated sludge process an F/M ratio approximately
equal to 0.5 results in a flocculated biological growth. BOD removal and oxidation are obviously
affected by such growth. The amount of oxidation depends on diffusion of oxygen into the
biological floe, where it is subsequently utilized by the microorganisms. At conventional mixing
intensities, relatively large floes are generated. The portion of these floes that are aerobic depends
upon a balance between oxygen diffusion into the floe from the surrounding liquid and the
oxygen consumption by the organisms contained within the floe. At low temperature, a low
oxygen utilization rate permits diffusion of oxygen to a greater depth in the floe and therefore a
large portion of the floe is aerobic. At high temperatures, the increased respiration rate depletes
the oxygen rapidly, and only a small portion of the floe is aerobic. It can be postulated that a large
mass of organisms at a low respiration rate (winter) achieves the same degree of oxidation as a
small mass at a high respiration rate (summer), and hence the computed coefficient 6 may be close
to 1.0. At high mixing intensities in the aeration basin, the smaller floe sizes may be fully aerobic
under all temperature conditions and the coefficient 8 will increase. At the higher F/M ratios,
filamentous or dispersed growths will exhibit a high temperature dependency as shown by
Wuhrmann (1966).
Process
Stabilization Pond
Activated Sludge
Aerated Lagoon
Trickling Filter
Aerobic - Facultative Lagoon
Anaerobic Lagoon
Extended Aeration
* This paper
TABLE 3
Summary Table
9 Range
1.072 - 1.085
1.0 -1.041
1.026-1.058
1.035
1.06-1.18
1.08 -1.10
1.037
Temperature
Range ° C
3 -35
4 -45
2-30
10-35
4 -30
5-30
10-30
Reference
Rowland (1958)
Eckenfelder (1970)
Dietzetal. (1966)
National Sanitation
Foundation (1966)
-------
189
It would therefore appear that the effect of temperature on the activated sludge process is related
both to the loading level (F/M) and to the intensity of mixing in the process. Trickling filters are
analogous to activated sludge except that oxygen diffusion into the film is uniplaner which results
in a 6 value in the order of 1.035.
Waste stabilization ponds, aerated and aerobic-facultative lagoon and anaerobic pond processes
operate at lower solids levels and/or higher F/M. Recent data have shown that the biomass does
not effectively flocculate at concentration levels below approximately 500 mg/l. Therefore, within
tfiese systems, the biomass is dispersed and hence the process is more directly affected by
variations in temperature.
High solids concentrations and long retention periods contribute to a low F/M ratio
(approximately 0.12) and a dispersed floe in the extended aeration process. The value of & would
therefore be expected to be higher than that for conventional activated sludge.
The results for various processes are summarized in Table 3.
CONCLUSIONS
1. The effect of temperature upon process efficiency can be formulated in terms of a temperature
coefficient, 6. 6 can be obtained graphically by variations of the modified Van't Hoff-Arrhenius
equation.
2. Temperature effects are minimal on activated sludge process performance as compared to
other treatment systems. The low sensitivity of the activated sludge process is primarily due to the
F/M ratio which results in a flocculated biological growth.
3. There is some evidence that 6 will increase at temperatures less than 10° C.
REFERENCES
Carpenter, W. L, Vamvakias, J. G. and Gellman, I. (1968) Temperature Relationships in Aerobic
Treatment and Disposal of Pulp and Paper Wastes, J. Wat. Pollut. Control Fed., 40, 783.
Hermann, E. R. and Gloyna, E. F. (1958) Waste Stabilization Ponds, III, Formulation of Design
Equations, Sewage Ind. Wastes, 30,963.
Hunter, J. V., Gentefli, E. J. and Gilwood, M. E. (1967) Temperature and Retention Time
Relationships in the Activated Sludge Process, Proc. 21st Purdue Ind. Waste Conf., 121, 953.
Oswald, W. J. (1966) Advances in Anaerobic Pond Systems Design, Advances in Water Quality
Improvement, Univ. of Texas Press, Austin.
Sawyer, C. N. (1966) New Concepts in Aerated Lagoon Design and Operation. Advances in Water
Quality Improvement, Univ. of Texas Press, Austin.
-------
190
Suwannakarn, V. and Gloyna, E, F. (1964) Efect de la Temperature en el Tratamiento de Aquas
Residuales Mediante Estanques de Estabilizacion, Vol. de la Off. Sanitaria Panamericana,
World Health Org., 43,128.
Wuhrmann, K. (1966) Research Developments in Regard to Concept and Base Values of the
Activated Sludge System, Advances in Water Quality Improvement, Univ. of Texas Press,
Austin.
-------
EVALUATION OF AERATED LAGOONS AS A SEWAGE TREATMENT
FACILITY IN THE CANADIAN PRAIRIE PROVINCES
Archie R. Pick, George E. Burns,
Dick W. Van Es and Richard M. Girling
INTRODUCTION
Metropolitan Winnipeg has a population of 500,000 and is located at latitude 49° 45' N, longitude
97° 15' W. The climate is of the continental type, with an annual temperature of 36.50° F, the
coldest month is January with an average temperature of -20° F.the warmest month is July with
an average temperature of 67° F above, and the average frost-free period (32° F) is 115 days. The
average winter snowfall is 51 inches.
In 1967, the Metropolitan Corporation of Greater Winnipeg undertook a two-year study of aerated
lagoons, because there was little documented information on aerated lagoons operating under
Canadian prairie conditions. In order to assess the applicability of this process for the treatment of
domestic wastes, three pilot aerobic-anaerobic aerated lagoons were constructed by the
Corporation.
During the summer and fall of 1967 the pilot lagoons were constructed in the corner of an existing
stabilization pond. The sewage treated was domestic sewage from a separate system. The three
aeration systems installed were:
Air-Aqua*
Mechanical Surface Aerator**
Air-Gun***
The work was supported by Public Health Research Grant 606-7-167 of the Department of
National Health and Welfare, Canada.
DESCRIPTION OF THE PILOT LAGOONS
Each system was designed to treat a flow of 0.5 Imgd. The general arrangement of the systems is
shown on Figure 1 and the design data is summarized in Table 1.
Air-Aqua
This system operates on the diffused air principle. A 30-HP compressor supplies air to
polyethylene tubes laid in a tapered grid on the cell bottom. The system has a 30-day retention
time, an operating depth of 10 feet and is divided into two cells, operating in series. 10% of the
effluent is returned to the inlet for seed.
* As manufactured by Hinde Manufacturing Limited, Hamilton, Ontario.
** Equipped with Lightnin Aerators as manufactured by Greey Mixing Equipment Limited,
Toronto, Ontario.
*** As manufactured by Aero-Hydraulics Corporation, Montreal, P.Q.
191
-------
192
MOTE
ALL DIMENSIONS TAKEN
AT BASIN FLOOR
•I'-IO"
SURFACE AERATOR
FIGURE 1 Plan of the aerated lagoons
-------
193
TABLE 1
Summary of Design Data
Item
Average Design Flow
Influent 5-day BOD 20° C
Influent Suspended Solids
BOD Removal rate Coefficient (Base 10)
k, atO C
BOD Removal rate Coefficient (Base 10)
kj at 20 C
Temperature Coefficient 8 (applied to k]
Oxygen Utilization factor a (Ibs.
oxygen required per Ib. 5Tday
BOD removed)
Oxygen transfer ratio (ratio of O2
transfer to waste to that of water)
(a factor)
Saturation value of waste compared to
H2 O ((3 factor)
Solubility of oxygen 20° C (780')
Operating dissolved oxygen
Effluent Temperature - winter
Effluent Temperature - summer
Influent Temperature - winter
Influent Temperature - summer
Mean ambient air temperature - winter
Mean ambient air temperature - summer
Treatment efficiency required
Retention time - Air-Aqua
Retention time - Surface Aerator
Retention time - Air Gun
Operating depth - Air-Aqua
Operating depth - Surface Aerator
Operating depth - Air-Gun
Volume - Air-Aqua
Volume - Surface Aerator
Volume - Air-Gun
Mixing requirements for surface aerators
Process loading - a) Design
Process loading - - Air-Aqua
Process loading - - Surface Aerator
Process loading - - Air-Gun
Process loading -b) Actual
Process loading - - Air-Aqua
Process loading - - Surface Aerator
Process loading - - Air-Gun
Design Formulation:
Value and Units
0.5 Imgd
250 mg/1
180mg/l
0.13 per day
0.50 per day
1.072 per ° C
1.50
0.85
0.95
9.02 mg/1
2.00 mg/1
32° F
75° F
48° F
65° F
-16°F
+75° F
90%
30 days
20 days
20 days
10ft.
lift.
17ft.
15 x 10* gal.
10 x 10° gal.
10 x 10° gal.
0.016 HP/1000 gals.
0.52 Ibs. BODs/1000 ft3/day
0.78 Ibs. BODj/1000 ft3, /day
0.78 Ibs. BODS /1000 ft3 /day
0.37 Ibs. BODj/lOOO ft3/day
0.55 Ibs BOD5/1000 ftyday
0.55 fbs. BODS/1000 ft*/day
2.3kj (100 -E)
Ibs. O2 per day =
= a' Ibs. BODS removed/day
-------
194
Surface Aerator
This system consists of eight 20-HP aerators installed in series along the lagoon basin. The basin is
sized for a 20-day retention time and has an operating depth of 11 feet. The raw sewage is fed in
adjacent to the first aerator only.
Air-Gun
A combination of the diffused air and surface aerator principles is used for this system. There are
54 guns installed in a tapered grid pattern. A 40-HP compressor delivers air into an inverted siphon
in the gun base, where a large bubble is formed which rises inside the gun, pushing water ahead and
exploding at the surface. The system has a 20-day retention time and an operating depth of 17
feet.
RESULTS
Effluent Quality
The raw sewage concentration for the period January 1, 1968 to September 30, 1969 averaged
175 mg/l and 188 mg/l for BODS and SS respectively. Corresponding effluent quality for the same
21 month period was:
BOD rag/1 S.S. mg/l
Air-Aqua 37 34
Surface Aerator 38 39
Air-Gun 34 34
Figures 2 and 3 show these in detail.
The effect of cold weather on BOD5 removal is evident when the winter of 1968-69 is compared
to the summer of 1968 on Figure 2. The loss of efficiency during the summer of 1969 is attributed
to sludge which will be discussed below.
The results were analyzed statistically for the 12 month period September 30, 1968 to September
30, 1969 (See Fig. 4 & 5). This period was selected as the most typical for continuous operation.
The median value of the effluent BOD for the three systems ranged between 39 and 38 mg/l. On
10% of the occasions the effluent was greater than or equal to 78 mg/l.
Suspended Solids removal remained reasonably constant over the 21-month period, with the
exception of start-up, which can be attributed to initial erosion and suspension of materials from
construction. On the 12-month basis the median SS in the effluent ranged between 21 and 23 mg/l
and on 10% of the occasions greater than or equal to 48 mg/l.
Figures 6 and 7 illustrate results for temperature and D.O. respectively.
-------
195
260
240
§ 220
|j 200
EJ 180
a.
w 160
g 140
§ 100
=i 80
5
cc 60
UJ
Q_
«, 40
O
100
BO
H 60
^ 40
20
•— ^
- — i
— —
JAN
a
~ea
Lf
RAW-
AIR A
SURB
AERO
'
^==
—
MAR
-,-.
*c»
m
E_— 1«
MAY
1
_ R
- R
MM
fc^=
JUN
OOj OF RAW SEWAGE
or%
^m
H.U
OF
•
^^^
EFFLUENT
i^S
C-CB
=_-.i
REMOVAL OF BOC
JUL
AUG
SEP
1968
iSEND
QUA
\CE
HY
AERATC
DRAULJ(
A
s
OCT
NCV
DEC
3
JAN
"T
r^:
MAf
__
APR
r^
„.
4AY
L—
r-i
***"
=
JUN
L
- _.
_- -T:
"TT^
HJL
•••
— —
mmrm
AU6
^=
__
._.
SEP
1969
FIGURE 2 Lagoon performance (BOD5)
-------
196
o
_l
cc
U!
u
tr
UJ
a
£WJ
260
240
220
200
180
160
140
120
IS8
100
80
60
40
20
,08
80
60
40
20
— i
*_ — .
—
—
JAN
R
RM
i
c£3
— i
__.
MAR
SUSPENDED SOUDS OF_RAW SEWA
— 1
— -
*- ~
1
SPENDED SOLIDS OF EFFLUENT
Ml
- —
:K^
--,
— —
—
^^^
=HE
- — •
m
L-_JI
•«=M
% REMOVAL OF SUSPENDED SOLI
APR
M*
JUN
oU_
AUG
SEP
LEGEND
fluu
AIR AQUA
SURFACE
AERO W
AERATOR-
T)RAUIJCS
OCT
MCV
DEC
p=
~
nc
JAN
r^
n
FFB
^~™
i
E^
MVi
— —
E
APR
i 1
•^
MAr
JLM
— i
•Ui
JUL
^^B
^
^=J
AUG
L^ .
=-=T3
SEP
FIGURE 3 Lagoon performance (Suspended solids)
-------
197
Nutrients
Figures 8 and 9 show the average performance of the aeration systems in the treatment of algal
nutrients, phosphorus and nitrogen. The average percent removals over the test period were:
System Nutrient Removal
Total Nitrogen (N) Phosphate (PO4)
Air-Aqua 12.6% 16.9%
Surface Aerator 14.6% 19.5%
Air-Gun 10.0% 23.3%
Although there was a reduction of total phosphorus through all systems, the orthophosphate
concentrations in the effluent increased. The nutrient removal efficiencies were relatively low
compared to results on ponds reported by others (Azzenso and Reid, 1966). The conventional
activated sludge plant operated by the Corporation has removal rates of 36% and 45% for nitrogen
and phosphate respectively. There was no appreciable difference in removal rates between summer
and winter.
The aeration systems are basically equal in nutrient removal and are relatively ineffective.
Temperature
The effluent temperature follows the ambient temperature curve closely as it is almost
independent of the raw sewage temperature (Fig. 6). There is a four-month period during the year
when the effluent temperature is between 0° C and 1° C.
One of the concerns in design was the possible freezing of the Air-Gun cell, as little information
was available on heat loss through ice cover. To keep the heat loss at a minimum, the surface area
was reduced by making the side slopes steeper. Observations proved, however, that the cell had a
built-in self-protection system. When the temperature rose to 10° F the ice melted on 25% of the
cell, but as soon as the temperature dropped the cell covered with ice to conserve heat.
SLUDGE ACCUMULATION
The three aeration systems have shown a substantial build-up of bottom sludges. Samplings were
conducted during July 1968, in the fall of 1969 and early in 1970. The accumulation of sludge in
aerated lagoons has been recognized by others (Thimsen, 1965; Barnhart, 1965; Clark and Dostal,
1968); however, the significance of the accumulation of sludge deposits under climatic conditions
similar to those experienced in the Canadian prairies has not been reported.
The rate of sludge accumulation in the Air-Aqua system has been estimated to be approximately
0.18 to 0.25 Ibs. of dry solids per capita per day (based on population equivalents). The rate is
comparable to the 0.21 Ibs. of dry solids per capita per day pumped to digestion at an activated
sludge plant operation by the Corporation.
Theoretical Sludge Accumulation
Theoretically, the overall digestion rate is sufficient to reduce the amount of volatile sludge
accumulated to a relatively stable content each summer. After the first summer of operation an
-------
198
BC
70
I
1 T ' i
LEGEND
SURFACE AERATOR
AERO HYDRAULICS-
AIR AQUA
PERIOD — OCT. 68-
"1
NO. OF
2
0
OCT.
I
X
2
69
^
*' S
^
TES"
53
B
51
•
•
'
'
•
/
s
7
/
IX
X
/
/
>'/
/
//
2
•
£
'•'
r
_/j.
I
1
t
'
X"
50
40
30
-
01 06 2 5 I 2 345 10 15 20 30 40506070 608590 9596979699 998 9 96 9999
PERCENT OF TIME EFFLUENT LESS THAN
FIGURE 4 Variability of final effluent BODS (Test lagoons on a one year basis)
annual cycle should occur with the sludge accumulation reaching a maximum in March-April and
decreasing to a minimum level in August-September. The theoretical cycle is shown in Figure 10
(based on temperature-corrected anaerobic digestion rates).
Based on the accumulation of all suspended solids removed, and 0.45 pounds of solids synthesized
per pound of BODS removed, the daily sludge production would be 1,120 pounds. If the end
products of anaerobic digestion had been removed from the systems, the amount of accumulation
at the September 1969 sampling would have been an estimated 200,000 pounds (average daily
accumulation of approximately 300 pounds).
Observed Sludge Accumulation
Based on actual surveys, the sludge accumulation (as dry solids) to September 5, 1969 was
estimated as follows:
Air-Aqua Primary
Air Aqua Secondary
Air-Gun
Total Lbs.
740,000
145,000
580,000
Lbs/day
1,120
220
930
Lbs/0/day
0.21
0.04
0.18
-------
199
fjj
IOU
ISO
14O
1 30
I20
••:
I AH
90
80
70
60
5C
40
3C
20
0
LEGEND
SURFACE AERATC
AERO HYDRAULK
AIR AQUA
PERIOD— OCT.
NO. OF Tl
>R 23;
•»q. irw
68
„ _
22
- OCT 69
c
---
;
0^*
•S\
-
~
^»
^
-*-^
^
&
^
*t
V
,-
.
&
f
/
j/
f*
..
*
•
t
X
/
/
^
X
y
X
./
,/"
//
7
>
-
^
**
^
.0! .05 .I .2 .5 I
2345 10 15 20 30 40 50 60 70 80SB 90 9596979899
PERCENT OF TIME EFFLUENT LESS THAN
99.8 9 93 9999
FIGURE 5 Variability of final effluent suspended solids (Test lagoons on a one year basis)
Sampling of the Air Gun had to be done through 15 feet of supernatant liquor and these figures
are subject to confirmation. Only two samples were obtained in the Air-Aqua secondary, so the
reliability of the 145,000 pound estimate is poor. Considering the methods of obtaining data, it
was concluded that there was no apparent difference in the quantities of sludge between the
systems, the significant point being that the accumulation was considerably greater than expected.
An analysis of the sludge from the Air-Aqua system indicated a moisture content of 90% for the
primary cell and 92% for the secondary cell; the volatile content was 55% for the primary and 45%
for the secondary.
Effect of Sludge Accumulation on Aerated Lagoon Performance
During the months of May, June, July and August 1969, in all three systems, there were significant
upward trends in the effluent BOD, as shown in Figure 11. All three systems showed similar
relationships. This decline in BOD removal efficiency was accompanied by a trend to reduced
dissolved oxygen concentrations in all cells. The reduced dissolved oxygen could be an "effect"
caused by high oxygen demands imposed by the end-products of anaerobic decomposition of the
sludge, or it could be a "cause" of higher effluent BOD due to an oxygen deficiency.
-------
200
Some of the related factors that may account for the sludge accumulation and the effect on
performance are:
a) Release of sludge digestion end products to the mixed liquor
b> Recycle of sludge by bacterial and algal synthesis in the mixed liquor
c) Relatively short period of higher rate anaerobic digestion
d) Insufficient air supply
The duration of the study has been insufficient to allow definition of the ultimate extent of sludge
accumulation and loss of efficiency during the summer. However, based on the observations made,
it appears that sludge is accumulating at the rate of approximately one ton of dry solids per Imgd
treated.
To determine if an abnormal quantity of inert material was contained in the raw sewage or
effluent a series of tests was conducted. The volatile content of the raw sewage was normal and
showed no evidence of extraneous inert matter. Similarly, the effluent volatile solids were typical
for biologically treated sewage.
Insofar as the higher BOD and reduced D.O. in the effluent is concerned, it is probably that
_
-
-
20
19
18
17
6
!
-
:
I?
II
10
9
8
7
e
•
3
2
:
^
^^--
^—-,
I
,
1
I
/
,'
____
/
/
f^**"
f
:
T
1
1
1
1
1
>
X
*/
& I
* !
i
I
,
~T7
<£/ -
g t
f ~B
i >
//
•f
^ j
A/
st^
"^^
^-^
EFFLUE
AMBIEN
s " ~».
\
\
\
J^"
, "
"\_
^z
r — - — •
HI
\
\
k \
^ \
\ ^>
\
\
v
\
\
i
W.,
\
\
\
\
s
\
\
V
V
\
\
\
\
^^-^
^
\
\
\
\
\
••-
\
\
**^.^
20
18
16
14
12
10
e.
e
4
2
0
-6
-e
- :
- A
16
16
-?n
a
'
J A
1968
FIGURE 6 Temperature monthly average
-------
201
\
/
/
\
„,"
^
1
1
\
\
\
— % —
\
\
\
\
\
W
\l
I
l\
1
\
\
\
\
\\
\\
\
^
r- —
/
/
/
//
If
t
/
\
\
\
\
\
/
/
/
/
/
/
/
/
/
/
X
\
\
^\._
\
•
X
K
\
\
SURFACE /
AERO HYD
AIR AQUA
X
, — •
\
\
V
\
V
-——
AERATOR- -
RAULICS—
, — -*~
,--
\
\
\
\
'
\
\
\
\
\
\
\
FMAMJ J AS ONDJ FMAMJ JAS
FIGURE 7 Dissolved oxygen average monthly (372 readings per system)
sufficient additional air can be added economically to handle the benthal demand, although some
doubts exist because the Surface Aerator system sustained a similar loss of efficiency in spite of
relatively higher D.O. concentrations. The major problem would appear to be the physical
accumulation of the sludge, which will ultimately require removal and additional treatment.
Sawyer has described the problems encountered as being similar to experience with Imhoff tanks,
with respect to removal and storage of BOD and solids during the winter months and the release of
soluble BOD and nutrients as acid fermentation of accumulation sludge deposits develops.
(Sawyer, 1970).
OPERATIONAL OBSERVATIONS AND PROBLEMS
Observations were made on mosquitoes, weeds, grease and scum odors, ice, foaming, grit, erosion
and equipment operation. Plugging of the Air-Aqua tubing and the Air-Guns occurred. The
problems were corrected and at the time of writing, both systems appear to be operating
satisfactorily, although more time is needed to assess their ultimate reliability. Ice was a major
problem with operation of the surface aerators; ice built up on the impeller shaft and support
structure with subsequent freeze-up and stoppage. Although steps were taken to try and eliminate
the ice problem, it was found that during the winter months only half of the surface aerators could
be kept operating.
-------
202
40
30
20
0>
fee
tv
8
Q_
10
a: to
(T
J
o LJ
23
-------
203
cell which at times released odors that were quite noticeable on the enclosing banks. They were
attributed to the floating grease and scum collecting along the banks and in the corners of this cell,
where this material would decompose, releasing odors. This occurred primarily along the south
banks, and may well have been the result of the very flat slope (8:1 > incorporated in the design.
Ice
Ice build-up caused the greatest problems with the surface aeration system. The Air-Aqua and
Air-Gun systems were notably free of problems due to ice build-up.
The Air-Aqua system did not cover completely with ice for the winter period and showed the
peculiar build-up of "stooks" over the air lines as observed in other installations. Ice thickness
found on the Air-Aqua secondary cell varied from approximately 6 inches to 48 inches from inlet
to outlet, respectively. With water depth of 10 feet this will cut down the retention period
considerably during the winter when water temperatures are adverse to the promotion of
biological activity. The foregoing is of course true for all lagoon type systems, be they
conventional or aerated. No ice thickness surveys were carried out on the remainder of the test
cells due to thin ice.
The Air-Gun system retained open water longer than the Air-Aqua. With the onset of colder
temperatures the openings immediately above the guns decreased in size from inlet to outlet.
Under severe temperature conditions the openings became covered with ice domes towards the
outlet end. Open water conditions existed year-round above the first two or three rows of guns.
40
30
i-
UJ
< iti
>.
N02 * N03
AMMONIA
ORGANIC
10
211 211 210 186 NO. OF TESTS
FIGURE 9 Test aerated lagoons average test results nitrogen
-------
204
Ice built up on the surface aerators to such an extent that it was impossible to clear the ice from
the aerator area. The result was that during the winter only three or four surface aerators could be
kept operating.
Foaming
As with conventional lagoons or stabilization ponds, foaming conditions were encountered to
some degree with all three aeration systems. The surface aeration system showed more foam than
the other two systems, due to the more violent agitation of the surface of the water by the
aerators.
The quantity of foam generated varied with the water temperature, the maximum condition
occurring after spring break-up when water temperatures were rising. However, the foaming did
not reach a point where it became a problem
JND SETTLED PER DAY)
LJJULJ
o '-
Q.
$ 160
u.
Q
3 I2°
Q.
«
O
o 40
S
0
/
~f
£
'
^-*
•- •
-^
X 680
-v
^_^_^
IHH
SL
^
-— ^
RET 1C
DGE
ATH"
/
,--^'
AL>
*
^
VOLJ
^-1
INE
N
kTILE
-^-"
^T
V
'
BYS
^
^-^
URVE
^
/
---
246 8 10 12 14 16 18 20 22 2
YEAR I YEAR 2
/
/
^~^~
/^
^,
\
\
^
^\
•^
^
- -
/
^-
4 26 28 30 32 34 3<
YEAR 3
FIGURE 10 Theoretical sludge accumulation in aerated lagoon (Complete anaerobic digestion
assumed)
Grit and Rags
After one year of operation considerable quantities of grit and rags were present in all three
systems. For the effective long-term operation of any aerated system pre-treatment facilities will
be required to remove grit and rags.
Bank Slopes
The following observations on bank slopes were made:
-------
205
The existing 8:1 slope utilized in the Air-Aqua cells created a wide unaerated band around a
portion of the lagoon; this likely resulted in some short circuiting. If flat slopes are required for
stability, consideration should be given to providing aeration along the slope. The majority of the
dikes for the demonstration lagoons were constructed at 3:1; this slope was stable, but for
long-term operation 4:1 is recommended. The slopes of the Air-Gun cell were constructed at 2.5:1
to minimize surface area. These slopes proved unstable under draw-down conditions. The rip-rap
provided effective erosion control in all cells.
Equipment Operations
Air-Aqua System. After start-up, difficulties were encountered with pressure build-up in the
system. Recommended HCi acid gas cleaning of the tube system was performed without alleviating
the problem. Removal of a 250-foot length of tubing revealed that the perforations had not been
«*
o
ci
ci
C)
-------
206
properly cut. All tubing was repunched in situ, requiring the lowering of the system so that the
workers could walk through the cells feeding the tubing through a punching mechanism mounted
on a boat. The repunching took 2 man-days to complete.
Upon completion of this work, pressure on the system dropped to 5.6 to 6 psig. The manufacturer
then recommended that acid cleaning be carried out quarterly, regardless of pressure build-up.
When operating properly, the pattern of air distribution should show a grid of air bubbles, released
by the perforated tubing. Although this grid did show immediately after installation and
repunching of the tubing, it deteriorated after a period of operation due to clogging of the
perforations, regardless of acid cleaning, carried out regularly. This clogging had affected the air
distribution pattern and was confirmed by D.O. tests. Further investigations traced the loss of
pattern to water in the tubes and condensation, and from water coming back through the valves
during power failures. By manipulating the acid valves it was possible to force the majority of the
water out of the tubes, restoring a reasonably good air pattern.
Supporting equipment such as air-compressors, effluent recirculation pump, flow meters, and
samplers were subjected to a regular preventative maintenance program, and few problems were
encountered.
Surface Aeration System. Ice has been a problem with the operation of the surface aerators. The
basic problem was ice build-up on the impeller shaft and supporting structure, with subsequent
freeze-up and stopping of the mechanism. Ice formed on the piles and impeller, resulting in a
limited clearance between the rotating ice on the impeller and the fixed ice on the piles. This is a
problem that is difficult to overcome with the winter climate experienced in the Winnipeg area.
Two methods aimed at overcoming this problem were attempted; firstly, one of the platforms was
shrouded with plywood, and secondly a second unit was shrouded with flexible nylon cloth. Both
attempts failed.
Other results of the ice accumulation were off-balance, causing vibration of the supporting
structure and misalignment of motor reducer, resulting in a coupling failure and loosening of
impeller, bending and loss of blades.
In the spring of 1968 the manufacturer replaced all blades with a heavier design. Although this was
thought to cure the problem encountered with the impeller blades, subsequent winter operation
disproved this, as bending and loss of blades occurred.
It may be possible to reduce or eliminate the icing problems with a change in the supporting
structure. The existing structure with four piles provided a surface for the ice to grow on, a two
pile arrangement with the piles widely spaced may be more successful.
Air-Gun System. No problems were encountered with the operation of this system until October
1968. when it was noticed that the most northeasterly gun was discharging air continuously.
During the winter 1968-69 this condition spread to most guns. It was thought to be caused by the
build-up of either ice or rags in the syphon chamber. The latter proved to be true. In May 1969,
the manufacturer installed syphon chambers of a new design on 42 guns. The new design, having
larger clearances, may eliminate the problems of plugging of the syphon chambers with rags.
Insufficient time has elapsed since the modification to allow an assessment of their long-term
dependability.
-------
207
COSTS
Capital Costs
Two sets of cost data have been prepared for aerated lagoons; firstly, the 0.5 Imgd demonstration
lagoons and secondly, a general unit capacity cost curve.
All costs and estimates in this report are adjusted to an Engineering News Record Sewage
Treatment Plant cost index of 132.0. As costs change the appropriate adjustment must be made to
the reported costs.
The capital costs for the demonstration lagoons are shown in Table 2. These costs have been
arrived at by assuming each system would be as indicated in Figure 1, insofar as floor dimensions
and identical equipment layout are concerned. However, in order to relate realistic costs to a
permanent installation it is necessary to assume each unit is to be constructed as a totally
independent unit. An independent unit is one with independent piping, diking, influent-effluent
structures and electrical supply. Therefore, only items 1 and 2 of Table 2 are established directly
from the contract amounts. The remaining items were arrived at by combining estimated
quantities and the actual unit prices tendered. The general contractor for the lagoon project was
consulted on the tendered prices. All costs were considered realistic with the exception of rip-rap.
In the case of rip-rap, the contractor was of the opinion that the price should be 1.5 times the
tendered price (i.e. $12.00/cu. yd. in place).
The estimated quantities for earthwork were calculated on the basis of 3 feet freeboard, 12 feet
roadway width, and 4:1 dike slopes, rip-rap facing the interior dikes from toe of slope to 1 foot
above the normal operating level. These adjustments are considered necessary to insure a
permanent and relatively maintenance-free structure with the soils encountered in the Winnipeg
area.
Sodding and seeding quantities include sodding the interior dikes above the rip-rap and 50% of the
total dike crest, and seeding the exterior dike slopes. Roadway quantities are based on an asphaltic
surface treatment being applied to the unsodded dike crest.
The electrical costs include service entrance equipment, motor starters, and lighting for the
equipment and structures associated with the respective systems. Power distribution costs were
based on the billing received from Manitoba Hydro for the supply and erection of equipment, and
allocation to the respective systems on the basis of length of cable and number of poles required
for each.
Actual influent and effluent chamber and piping costs were reestimated on the basis of piping and
chambers being of such length and size for an independent treatment system of 0.5 Imgd capacity.
Fencing costs are for the perimeter of the cell taken at the toe of the dike slope. Additional items
of chlorination facilities and instrumentation are included since these would be desirable in a
permanent installation. The costs do not include pumping station, forcemain, outfall, or land.
However, they are inclusive of engineering, legal and administration charges.
From Table 2 it can be seen that the total costs for the Air-Gun are slightly lower than the
Air-Aqua and Surface Aerator Systems. Major cost differences are due to additional rip-rap and
-------
208
TABLE 2
Capital Costs - Demonstration Lagoons
Surface
Air-Aqua Aerator Air-Gun
1. Aeration Equipment, Supply
(Blowers, Headers) $ 47,500 $ 44,300 $ 38.0001
2. Aeration Equipment, Installation
(Headers, Housing, Platforms) 23,000 56,500 33,700^
3. Electrical 3,500 7,800 3,500
4. Influent Piping 3,200 3,200 3,200
5. Effluent Piping 3,500 2,200 2,200
6. Influent Chamber 17,900 17,900 17,900
7. Effluent Chamber 22,900 22,900 22,900
8. Excavation & Dike Construction 32,400 29,200 27,000
9. Clearing, Grubbing Unsuitable Material 10,800 10,800 10,800
10. Rip-rap 93,000 73,500 70,300
11. Seeding & Sodding 6,400 5,000 4,000
12. Roadway (asphalt) 3,900 3,500 3,200
13. Chlorination Facilities (including
housing) 20,000 20,000 20,000
14. Power (hydro) 3,500 5,400 3,500
15. Instrumentation (including magnetic
flow meter) 10,000 10,000 10,000
16. Fencing 4,700 4,000 3,500
Total Estimated Cost $306,200 $316,200 $273,700
10% Eng. & Contingencies 30,600 31,600 27,300
Total Estimated Cost $336,800 $347,800 $301,000
Costs indexed to ENR S.T.P. 1) Includes installation, excludes blowers and header
il
cost index 132.0 2) Includes blower and header
excavation required for the dividing dike, greater dike perimeter for the Air-Aqua System, and
platform costs for the surface aerator.
Figure 12 shows the development cost of the aerated lagoon process per unit of lagoon capacity.
These costs have been developed by applying the unit material and installation costs of the
demonstration lagoons to calculated quantities. Equipment supply and installation costs were
derived from equipment manufacturers' quotations for similar facilities at other treatment works.
-------
209
N
I
(fl
I/)
5
a
rrVQTQ t
INCLUDED
NOT
X — CAl
0.5
30
16 0
iN.R
10% E
ON A
OUTF
AND
_CULA"
IMGO-
IMGD-
IMGD-
— 1 — r
1 1 1
STP INDEX
NGINEERINI
:TUAL cos
ALL, LAND
BLOWER H
FED ESTIM
-1 CELL S
-2 CELL S
—3 CELL S
. ^
\ \ \
\ \
5 AND CONTINGENC
T, LEGAL, CHLORIN
, PUMPING STATIOf
OUSING
MES
YSTEM
YSTEM
YSTEM
S*
^
:IES
AT ION
1
— >
s*
^—
^,s
^
/
-
' T
0.1 2 .3 4 .5.67.8.91.0 2 346 678910 20 30 405060708090100
DESIGN CAPACITY — AVERAGE FLOW [I.M. 6.D.)
FIGURE 12 Aerated lagoons capital cost vs. design capacity
Operation and Maintenance Costs
The operating and maintenance costs for the 0.5 Imgd demonstration lagoons are shown on Table
3. They are based on 17 months of operation.
Assuming chlorination of the effluent at 8 mg/l is desirable, the total operating costs should be
adjusted to $53, $107 and $56 for the respective lagoons.
The power costs tabulated in Table 3 are based on field measurements of true power drawn with a
unit power cost of 0.9c/kwh. Allocated to each system is a shore of the heating and sampling
pump load. In the case of surface aeration the cost is calculated on all eight units operating. In
terms of oxygen requirements, the surface aerators are capable of treating in excess of the rated
flow. However, at the time of design, eight units were considered necessary to insure adequate
mixing.
Repairs include regular weekly inspection and servicing of equipment, non-routine repairs and
replacement parts where required. The costs of repunching the aeration tubes and replacing the
siphons on the Air-Guns are not included in Table 3. The labor and materials on the servicing were
based on actual time card and material receipts and allocated to the respective systems. Labor
includes a 25% factor for overhead. General costs include maintenance to the common inlet
chamber, metering equipment, etc. Since it was difficult to allocate these costs to a specific lagoon
they were split equally to the three systems. Laboratory costs include labor and supplies for
-------
210
TABLE 3
Actual Operating and Maintenance Costs
0.5 Imgd Demonstration Lagoons
Avg. Cost per Img Treated
Surface
Air-Aqua Aerator Air-Gun
Power $ 10 $ 64 $ 15
Equipment Servicing & Repair 13 14* 12
Laboratory & Control 13 13 13
Road & Dike Maintenance 433
Snow Removal, Grass Cutting,
Mosquito Control 333
General 333
Total Operating & Maintenance Costs
per Imgd Treated $ 46 $100 $ 49
*Not continuously maintained due to ice.
sampling and analysis. Actual laboratory costs were three times the cost reported due to the
frequency and number of analyses performed under test conditions. Therefore, the lower costing
for laboratory reported here would be a more realistic value under normal operating conditions.
Maintenance and laboratory personnel were not based at the lagoons during the test; therefore
considerable labor and vehicle time for travelling were not charged to the test lagoons.
The development of operating and maintenance costs for aerated lagoons is shown in Figure 13.
The curve was prepared by projecting the actual demonstration lagoon costs on the trend line as
indicated. The slope of the line was guided by costs reported in recent literature (Okey and
Rickles, 1968). The reported costs from the literature reference have been adjusted to Imperial
Gallons and are also shown on Figure 13. It is noteworthy that the referenced costs are
representative of power costs of 1 c per kw. hr. and labor of $5.00/hr., while in comparison the
local power and labor costs are approximately 0.9 c per kw. hr., and $4.25 per hr., respectively.
Very little cost information at other aerated lagoons is available to confirm the trend costs for
increasing capacities. However, a check was made on the expected operating and maintenance
costs by determining the future power requirement for mechanical equipment coupled with
current power rates. The total costs were extrapolated from the power costs on the basis of power
being about one-third of the total costs. This provided a reasonable check on the trend developed
by the literature.
CONCLUSIONS
Aerated lagoons were found to be capable of providing "secondary equivalent" sewage treatment.
Under prairie climatic conditions there is a problem of sludge accumulation leading to a decline in
efficiency of BOD removal and a reduction in the dissolved oxygen concentration during the
summer months.
-------
211
The economic feasibility of aerated lagoons is questionable until the extent and cost implications
of the sludge problem are fully defined by further research and experience.
The use of surface aerators under prairie winter conditions is not practical due to ice build-up
ft can be concluded that aerated lagoons are an effective means of providing secondary treatment
but some provision must be made for sludge handling.
It is intended to continue the investigation on a less rigorous scale, with a goal of determining the
long term effects, and finding a practical and economical solution. Equipment manufacturers are
actively pursuing a solution.
500
400
300
S 20°
Si
IE
H
* '38
£ %
60
M
1
<2 40
§ 30
1 »
10
~--~.
REPOf
,^cos-
'~ ^_
TED R
S (RE
ANGE
f. 10)
or
S
1
M
*2a
~--^
COSTS (E.N.R. STP INDEX I32J
INCLUDED. (I) POWER, EQUIPMENT AND PARTS
(2) LABOR (INCLUDES 25% FOR 0V
(3)SNOW REMOVAL, GRASS CUTTINf
INSECT CONTROL, AND ROAD MA
(4) LAB ORATORY
(MCHLOftlNATION
(6)5% FOR DIKES ROAD MAINTENCE
OTHERS NOT ENCOUNTERED OUF
(7) PUMPING
NOT
UdOJJBED' TRAVEL TIME, a VEHICLE COST F(
MAINTENANCE STAFF AND LAB.
11 ii |
'---.
b^e
ssfiai
ERHEA
INTEN
AND
ING TE
)R —
"--
59e?r
-».^
^
D)
Vk.
:ST t
A
^
'I K
.
frn
t'!\>
OD
1
-
I
^
i-'J
fcS
.3 A .5 6 .7 .8 .9 ID 2 3 4 5 6 7 8 9 10 20 30405060708090
DESIGN CAPACITY—AVERAGE FLOW (I.M.G.D.)
FIGURE 13 Aerated lagoons operating and maintenance costs
-------
212
REFERENCES
Azzenso, J. R. and Reid, G. W., Removing nitrogen and phosphorus by bio-oxidation ponds in
centra! Oklahoma, Water and Sewage Works, V. 113, No. 8, p. 294-299.
Barn hart, E. L. and Eckenfelder, W. W., Jr. (1965) Theoretical aspects of aerated lagoon design,
paper presented at the Symposium on Waste-water Treatment for Small Municipalities, Ecole
Polytechnique, Montreal, pp. 1.
Clark and Oostal (1968) Evaluation of waste treatment Chemawa Indian School, Report No. FR-6,
FWPCA, North West Region, Pacific North West Water Laboratory. Corvallis, Oregon.
Okey, R. W. and Rickles, R. N. (1968) Industrial waste treatment management. Water and Waste
Engineering, Vol. 5, Nr. 9, pp. WWE/1-14.
Sawyer, C. N. (1970) Private communication.
Thimsen, D. J. (1965) Biological treatment in aerated lagoons. Paper presented at the Twelfth
Annual Waste Engineering Conference, University of Minnesota.
-------
DESIGN CONSIDERATIONS FOR EXTENDED AERATION IN ALASKA
Sidney E. Clark, Harold J. Coutts and Conrad Christiansen
INTRODUCTION
Alaska is the largest state of the United States, sparsely populated and with a variety of climates,
including arctic, sub-arctic, marine sub-arctic and temperate. The population is small and
widespread with 294,417 people (preliminary 1970 census figure) inhabiting 586,000 square miles
of land area. Settled areas requiring domestic sewage treatment include large municipalities,
military installations, remote sites and villages, each of these having different requirements and
presenting different problems.
Construction and power costs in Alaska are very high in general, and excessively so in remote areas
(1.5 to 5 times Seattle construction costs index). Skilled personnel for operation of treatment
plants are scarce and, in most cases, nonexistent.
The effect of man's waste on arctic and sub-arctic ecosystems has received little attention in the
past and is not well understood. Because of recent increased interest in the Arctic region, some
information is now becoming available on man's possible influence. For example, during the
winter, dissolved oxygen (DO) of ice-covered Alaska rivers may reach extremely low levels of 3
mg/l or less under natural conditions (Frey, 1969; Gordon, 1970; Roguski, personal
communication). Because of the retarded ability of Alaska streams to replenish DO under total or
nearly total ice cover, it becomes essential that the natural balance not be upset by man. Under
these conditions, secondary sewage treatment will be required to assure adequate stream
protection.
One of the major advantages of biological processes for provision of secondary treatment is their
ability to oxidize waste without large inputs of energy, thus reducing shipping costs, etc.,
associated with materials required for chemical treatment. All factors considered, extended
aeration systems have considerable potential for reliable and economical secondary treatment at
larger governmental installations and large communities in Alaska (populations greater than 250).
Current extended aeration research is being conducted by several groups:
1. Corps of Engineers
Cold Regions Research and Engineering Laboratory
Alaska District
2. University of Alaska
Institute of Water Resources
3. Federal Water Quality Administration, Alaska Water Laboratory
Cold Climate Research Program
213
-------
214
Waste treatment research at the Alaska Water Laboratory is concerned primarily with adapting
methods developed in the contiguous United States to the extreme cold climates found in Alaska.
The scope of the present work on activated sludge is, in general, limited to extended aeration, and
includes investigations in the following areas:
1. Low temperature biokinetics
2. Low temperature solids removal
3. Degree of environmental protection required for equipment and processes.
4. Aeration requirements
5. Aeration chamber mixing
6. Waste sludge characteristics and disposal
The above investigations are being conducted on a laboratory and pilot plant scale. Monitoring of
existing facilities is also taking place.
LITERATURE AND EXPERIENCE REVIEW
Low Temperature Biological Treatment Feasibility
Although the activated sludge process is affected by temperature, operation at temperatures
approaching freezing is feasible. A number of investigators have reported a considerable amount of
biological activity taking place at freezing temperatures and below (Ayres, 1962; Halvorson,
1962). Miller (1967) has reviewed the information available on microorganisms indigenous to cold
environments and found that research on psychrophilic organisms is still in the initial stage, but
concluded that "truly psychrophilic microorganisms do exist and are distinguished by their ability
to grow at very low temperatures and to do so at rates comparable to those of mesophiles at higher
temperatures." The feasibility of effective biological treatment by full-scale extended aeration
facilities at operating temperatures as low as 2° C has been demonstrated (Anonymous, 1965;
Grube and Murphy, 1968; Schmidtke, 1967).
Temperature Effects
Pasveer (1955) conducted laboratory scale temperature studies with activated sludge and reported
that the process goes on almost as well at 3° - 5° C as it does at 20° C. Wuhrmann (1956) found in
his studies of the activated sludge process that "the BOO removal seems to be only slightly
influenced by temperature, whereas nitrification is markedly higher in summer than in winter."
Ludzack (1965) conducted bench scale studies using a continuous apparatus with a detention time
of 24 hours and a loading of 35 Ib. COD/1,000 ft3 and demonstrated COD removal efficiencies of
<90% at 21° - 25° C. 90% at 10° C and 84% at 5° C. Hunter et at. (1966) conducted batch
operated laboratory scale studies on the effect of temperature and retention times on the activated
sludge process. At temperatures between 4° C and 45° C, they found little change of BOD or
suspended solids removal efficiencies. As the temperatures increased, they found less filamentous
growth and increased protozoa and rotifer populations. Grube and Murphy (1968) evaluated an
-------
215
oxidation ditch and found BOD removal efficiencies greater than 90% with liquid temperatures of
2° C, air temperatures ranging down to -40° C, and average detention times of 2.3 days. Influent
temperatures averaged 16.6° C with a minimum of 7.5° C. Gustaffson and Westbury (1965)
evaluated the activated sludge process for application at Kiruna, Sweden, and obtained 75% BOD
reduction with a 31/2 hour detention time system at 2.8° - 4.8° C and 2,700-3,500 mg/l MLSS.
Temperature Coefficient
The temperature coefficient, 0, is used in the relationship
ki/k2 = B (t, -ta)
to define the effect of temperature on biological activity. The values Iq and k2 refer to velocity
constants at temperatures t, and t2 respectively. The value of 6 indicates the extent of the
temperature effect on the biological activity. Use of this equation, known as the Arrhenius
relationship, to define the effect of temperature on wastewater and reaction rates, dates back to
Streeter and Phelps (1925) and Theriault (1927), who reported 6 values of 1.047 for domestic
wastewater and river water (Zanoi, 1969). Pohl (1967) concluded that 6 was dependent on the
mixed liquor concentrations: 9 = 1.038 at low MLSS and 1.000 at high MLSS. Benedict (1968)
conducted studies in the temperature range of 4° - 32° C and concluded 6 (0 = 1.078 @ 4° C) was
independent of loading when the loading rate did not exceed 0.53 Ibs BOD/day/lb/MLSS, but 6
increased as loadings above 0.53 were imposed. Eckenfelder (1967) suggested that0, based on
overall treatment efficiencies, was a function of the organic loading and reported 0 values for
activated sludge of 1.00 at low loadings and 1.02 at high loadings.
Solids Separation
Solids removal plays a very important part in the efficiency of the activated sludge treatment
process. The degree of sludge separation directly influences the quality of effluent from
wastewater treatment plants with higher concentrations of effluent solids contributing to high
effluent BOD. Reed and Murphy (1969) conducted an investigation on settling characteristics of
activated sludge at temperatures ranging from 1.1° to 23.4° C and found that the influence of
temperature on settling velocity decreased as the concentration increased. They developed an
equation for zone settling based on experimental data. They also suggested upflow sludge blanket
clarifiers as having greater potential for cold regions application. Benedict (1968) suggested that
the effect of sludge settleability on gross COD removal was magnified at low temperatures and as
the loading rate was increased.
Hansen (1967) reported on a method of solids separation which successfully employed shallow
depth sedimentation theory. The settling units consisted of small diameter tubes (1-inch) inclined
at 5° and 2-4 feet in length. Detention times were very short and backwashing was necessary for
removal of accumulated solids. Hansen (1967) also reported on the use of steeply inclined tubes
(60°) which permit solids deposited in the tubes to continuously slide down by gravity. A
secondary clarifier of a trickling filter plant was converted to a biological reactor and the steeply
inclined tubes utilized for solids separation which increased plant efficiency from 85% to more
than 95%. The effluent suspended solids averaged 70 mg/l varying from a low of 7 mg/l to a high
of 190 mg/l, which was comparable to that produced by a conventional clarifier of an extended
aeration plant of the same capacity (3000 gallons per day at 12-hour detention). Other reports are
-------
216
available which describe the use of tube settlers in water treatment and waste treatment solids
separation (Gulp, 1969; Senlecta and Hsiung, 1969).
Pohl (1970) investigated tube settlers in the laboratory and obtained the best results at room
temperature but found the tubes passing excessive collodial solids occasionally.
Design Parameters
Little information is available on biological treatment process design for temperatures less than 5°
C. Ludzack (1965) and Hunter, et al. (1966) observed that excess MLSS accumulation increased
with decreasing temperature. The cell yield (c) increases with increasing temperature because it is
believed a larger portion of BOD removed is utilized for energy at low temperatures than at high
temperatures (Stewart, 1964). -Since the rate of endogenous respiration is depressed at low
temperatures, the quantity of excess sludge produced is increased. Benedict (1968) reports values
for c and k (endogenous rate) at 4° C of 0.42 mg/mg CODr and 1.32% respectively.
Aeration
Eckenfelder and O'Connor (1961) stated that the temperature coefficient 6, when applied to
oxygen transfer efficiencies, has been reported to vary from 1.016 to 1.047 and that studies on
bubble aeration indicated a temperature coefficient of 1,02 applied. The effects of temperature on
stream reaeration have been studied under controlled experiments in the laboratory (Anonymous,
1961). A value for 6 of 1.0241 for the temperature range of 5° to 30° C was found. Black (1968)
described a procedure for evaluation of aeration devices and stated that a 6 value of 1.030 or
higher should be used for cold water.
LABORATORY STUDIES
During the past two years, three bench scale activated sludge reactors have been utilized for
kinetics and solids separation studies. The three units are illustrated in Figures 1, 2, and 3. The
systems have been operated as continuous flow systems with the feed being primary effluent
Mnr
Conwrtrtc Ctnn
FIGURE 1 Cone reactors (as manufactured by Pope Scientific)
-------
217
Reactor
Sid* Vi*«
FIGURE 2 AWL reactor (8.9 gal.)
brought to the laboratory from the Eielson primary sewage treatment plant. Routine analysis
included influent and effluent BOD and COD, mixed liquor and effluent suspended solids (SS) and
volatile suspended solids (VSS). Nutrient analyses of the influent and effluent samples were made
weekly and included ammonia, nitrite, nitrate, organic nitrogens, total phosphates, and
orthophosphates. A limited number of coliform counts were made on the influent and effluent.
Microscopic examinations of the reactor contents were made on an irregular basis at times when
apparent or suspected changes in the mixed liquor had taken place. The examinations consisted of
general observations on the relative quantities of protozoa present and the degree of activity. BOD,
COD, and solids analyses were done in accordance with Standard Methods procedures (1965).
Coliform counts were made by the membrane filter method as described in Standard Methods and
nutrient analyses were made in accordance with Federal Water Quality Administrations Standards
(1969).
The cone reactors (Fig. 1), when operated at 1.3° C and 6.5° C for long periods of time, showed
some interesting characteristics which are summarized in Tables 1 and 2. Both biological sludges
were relatively easy to establish.
Air Suppy
l-teTimer
jj[s ienoki*vav«*
24-hr
Timer
Recycle
Pump (Hotter)
Effluent Tank
2 Tube
Moisture
Trap
FIGURE 3 AWL reactor (12.45 gal.)
-------
218
The reactor runs started with the longest detention time first and the times decreased in
chronological order. The 1.3° C reactor took a considerable amount of time to establish a stable
system (more than 3 months). However, a good removal rate was obtained before the MLSS
stabilized. There was apparently little difference in the biological activity at the two temperatures,
but operation of the reactor at 6.5° C was more erratic.
Both reactors generally showed "auto-induced sludge wasting" in the same manner as the College
Utilities oxidation ditch described by Grube and Murphy (1968). The MLSS would build up to a
point and begin to pass solids for 1 or 2 days and then repeat the cycle. The cycle was repeated
within 2 to 3 weeks as opposed to the monthly occurrence reported by Grube and Murphy.
The reactors differed in their manner of passing solids, with the 1.3° C reactor genera My having a
much more turbid effluent and the 6.5° C reactor having a relatively clear effluent. Heavy solids
passed from the 6.5° C reactor by rising in the settling tube as a solid mass. As the concentrations
of solids in the mixed liquor increased, the level of solids in the settling tube would rise until
spilling over into the effluent tank. After passing an undetermined amount of solids, the cycle
would be repeated. A gradual drop in pH was noted in the 6.5° C unit as the suspended solids
began to build before discharging. The pH dropped from slightly above 7 to values of 6.6 to 6.7.
pH of the 1.3° C unit consistently remained around 7.4. The 6.5° C effl ent solids settled to the
bottom of the effluent tank leaving a clear liquid above, whereas, the 1.3° C effluent solids did not
settle to any degree. As the 1.3° C reactor became more stabilized, the effluent became less turbid
TABLE 1
Data Summary
1.3°CCone Reactor
Feed: Primary Plant Effluent
Detention
Time(hrs) 21 15 13 9
Influent
BOD(mg/l) 111 170 201 184
Reactor
Susp. Solids (mg/1) 1,074 1,561 2,657 2,926
Volatile
Susp. Solids (mg/1) 890 1,324 2,212 2,402
Filtered Effluent
BOD (mg/1) 37 11 20 14
% BOD Removal 66 93 90 92
Unfiltered Effluent
SUSP. Solids (mg/1) 29 43 38 82
BOD (mg/1) 40 62 28 44
% BOD Removal 64 64 86 76
Loading Factor
Ib BOD Fed
IbMLVSS-Day .19 .21 .17 .20
Product of
MLVSS and Det. Time 14,700 19,500 28,000 21,600
-------
219
and the MLSS began to increase. The 6.5° C reactor operation was less stable, with the maximum
level of MLSS generally not rising above 2,300 as opposed to 3,000 for the 1.3° C MLSS. Results
of nutrient analyses are presented in Table 3. There was a significant change in nitrate and total
nitrogen at 6.5° C when going from 9 to 13 hours detention time. This was also true at 1.3° C to a
lesser degree. There was a greater reduction in ammonia nitrogen and a greater increase in nitrate
nitrogen at 6.5° C. Ammonia was essentially not affected at 1.3° C. Total nitrogen removals were
much higher at 6.5° C than at 1.3° C with little detention time effects.
Overall results of operation of the 8.9 gal. and 12.45 gal. reactors are presented in Tables 4 and 5.
Temperature changes were accomplished by a gradual increase or decrease in the constant
temperature room temperature. These reactors were operated at 12-hour hydraulic detention times
with daily sludge wasting to maintain the MLSS at 4,000 mg/l. The 8.9 gal. reactor was later
converted to a 24-hour operation. Sludge was wasted by drawing off the required amount of
mixed liquor. A portion was used for a solids analysis to determine the exact amount of solids
removed. The effluent BOD and COD figures of 9 to 21 mg/l and 46 to 96 mg/l indicate that a
considerable amount of biological activity takes place at low operating temperatures.
Effluent BOD/COD ratios varied from 0.13 to 0.27, indicating that effluent organics were well
oxidized. These were in comparison with the influent BOD/COD ratios of 0.55 to 0.66.
TABLE 2
Data Summary
.0
6.5 C Cone Reactor
Feed: Primary Plant Effluent
Detention
Time(hrs) 17 15 13 9
Influent
BOD (mg/l) 139 132 153 155
Reactor
Susp. Solids (mg/l) 2,346 1,885 1,880 2,285
Volatile
Susp. Solids (mg/l) 1,915 1,563 1,587 1,801
Filtered Effluent
BOD (mg/l) 51.3 16.3 13.3 11.7
% BOD Removal 63 88 91 92
UnfUtered Effluent
Susp. Solids (mg/l) 11 69 96 45
BOD (mg/l) 53 36 31 33
% BOD Removal 62 73 80 79
Loading Factor
Ib. BOD Fed
Ib.MLVSS-Day .08 .106 .18 .23
product of
MLVSS and Det. Time 31,600 23,000 23,300 20,600
-------
220
TABLE 3
Cone Reactors Results of Nutrient Analysis
13 Hour Detention Time
1.3°C REACTOR
6.5°C REACTOR
Influent Filtered Unfiltered Influent Filtered Unffltered
Effluent Effluent Effluent Effluent
NH3-N (Ammonia)
NO2-N (Nitrite)
NO3-N (Nitrate)
KJeldahl-N (Nitrogen)
Total (Nitrogen)
Total Nitrogen
Removals (%)
O-PO4 (Ortho-Phosphate)
22
.13
.13
41
41.26
20
19
.09
2.13
28
30.22
27
18
18
.05
2.02
29
31.07
25
18
19
.11
.21
37
37.32
19
1
.13
9.17
3
12.30
67
18
1
.15
12.13
3
15.28
59
18
9
1.
Influent
Hour Detention Time
3°C REACTOR 6
Filtered
Effluent
Unfiltered
Effluent
Influent
NH3-N (Ammonia)
NO2-N (Nitrite)
NO3-n (Nitrate)
Kjeldahl-N (Nitrogen)
Total (Nitrogen)
Total Nitrogen
Removals (%)
O-PO4 (Ortho-Phosphate)
21
.06
.11
36
36.17
17
19
,,03
.68
26
26.71
26
14
19
.03
.54
27
27.57
24
15
6.5 C REACTOR
21
.06
.07
35
35.13
19
Filtered
Effluent
1
.14
8.03
3
11.17
68
18
Unfiltered
Effluent
1
.12
14.45
3
17.57
50
16
(1) Total nitrogen results reported are the sum of the nitrite and Kjeldahl nitrogen analysis.
The amounts of sludge wasted varied from 0.42 mg susp. solids/mg BOD removed at the low
temperatures to 0.14° to 10.5° C and 24-hour detention time. The pH of both reactors ranged
from 7.2 to 7.6 during the sample periods reported.
Poor settling sludges were developed during operation of these reactors with the Sludge Volume
Index (SVI) consistently ranging above 200. The sludge produced appeared to be of a zoogloeal
type similar to that reported by Heukelekian and Wiesburg (1965) who found a direct correlation
between increasing SVI and increasing bound water for this type of bulking. Very little evidence of
Sphaerotilus was noted during microscopic examination. Ludzack (1965) also reported a poor
settling sludge at low temperatures (5° C) with very poor drainability.
The significance of protozoa in an efficiently operating activated sludge process, as reported by
McKinney (1956), was observed during operation of the reactors even at the coldest temperatures.
The 12.45 gal. reactor was seeded at temperatures <2° C with return sludge from an oxidation
ditch treating domestic sewage. Initially, the effluent was very turbid as the sludge was acclimating
-------
221
TABLE 4
Summary of Results of 8.9 Gallon Reactor and
12.45 Gallon Reactor at 12-hr Detention
Feed: Primary Plant Effluent
Reactor MLSS (mg/1)
%VSS
BOD
COD
Loading:
Ib. Infl. BOD
lb MLVSS-Day
Sludge Wasted: mg/MLSS
rag BOD Removed
Unfiltered Effluent
Suspended Solids (mg/1)
BOD (mg/1)
BOD Removal (%)
COD (mg/1)
COD Removal (%)
BOD/COD Ratio
Influent
Effluent
Reactor
REACTOR TEMPERATURE (AVG°C)
4160
80
2489
5648
0.12
0.42
18
21
89
78
76
0.60
0.24
0.44
2.9
4097
80
2503
5788
0.10
0.33
3
13
92
46
73
0.60
0.23
0.43
3.8
4076
81
1477
5260
0.10
0.32
12
17
90
67
78
0.55
0.25
0.28
8.0
3737
80
1299
4705
0.14
0.33
5
9
96
96
83
0.66
0.16
0.28
itself to the new conditions. The decreasing turbidity of the sludge as acclimation progressed
corresponded to increasing numbers of protozoa, generally Paramecium and Vorticella. As
reported by McKinney (1956), a very well-stabilized activated sludge system will have few stalked
ciliates and no other protozoa because of relatively few bacteria, whereas, a somewhat less
stabilized system will have greater numbers of free-swimming ciliates because of greater numbers
of free-swimming bacteria. He stated that the presence of stalked ciliates indicates an activated
sludge system with a low BOD effluent. Vorticella was present in both reactors after initial startup
except for one period in the 12.45 gal. reactor as described below.
After stable operation at temperatures <2° C and ^4° C the 12.45 gal. reactor temperature was
increased to 8° C over a period of 6 days. The effluent suspended solids increased from
approximately 5 mg/l before the temperature increase to approximately 18 mg/1 during the
increase and reached a maximum of 46 mg/1 after 3 days at 8° C. During this period, the effluent
became turbid with few solids settling out in the effluent tank. The protozoa became very reduced
in numbers and inactive. Again, the return to normal operation corresponded to an increase in the
number of Vorticella and Paramecium present in the sludge. Coliform removal also corresponded
directly to the numbers of protozoa present, dropping from 99.8% removal before the upset to less
than 80% during the protozoa number reduction. Ten days after returning to stable operation at
8° C, the sludge was exhibiting the same characteristics as with the 8.9 gal. reactor. That is, the
-------
222
SVI was ranging around 250 and the floe exhibited a fluffy snowflake appearance. Operation of
the reactor was not impaired under these conditions because a backwash cycle was added to the
settling apparatus. Protozoa increased in numbers when the systems stabilized at 12° C.
Sludge wasting and disposal in cold climates should be given attention. Based on data presented
earlier, it would appear that provision should be made for wasting 0.5 Ib. solids per Ib. or BOD
removed at colder operating temperatures (<5° C) and at organic loadings of 0.1 Ib. influent BOD
per Ib. MLVSS-Day. Sludge digestion and disposal methods present a problem at colder
temperatures due to added heat requirements and poor drainability. Ludzack (1965) indicated
that sludge development at cold temperatures may require digestion at higher temperatures before
disposal. Thomas (1950) indicated the freeze-thaw cycle may be taken advantage of in cold
climates to increase drainability.
Tube settlers have been evaluated as a possible alternate means of providing solids separation and
return. During operation, sludge rises in the tubes until it reaches a level at which it is in
equilibrium with the effluent flow. Action in the tube consists of a rolling motion in which solids
are being carried up along the top side of the tube in a mass with the effluent, as shown in Figure
4. The mass gradually settles toward the bottom side of the tube where it enters a current moving
downward caused by the weight of the solids. During normal operation, solids in the tube are
constantly being replaced at a relatively high rate. In the temperature range of 0° through 4° C the
SVI of the mixed liquor ranged around 230 and did not hinder the operation of the reactor. At 8°
TABLE 5
Summary of Results of 8.9 Gallon Reactor
at 24-hr Detention
Feed: Primary Plant Effluent
REACTOR TEMPERATURE (AVG°C)
1.9 6.8 10.5
Reactor MLSS (mg/1) 2595 3872 3896
%VSS 83 83 82
BOD 1693 2105 1808
COD 3712 5019 5178
i™Mr,a- lb-Infl-BOD
Loading. _ OQ7 OQ7 Q07
lb MLVSS-Day
Sludge Wasted: mg/MLVSS 0.42 0,16 „ 14
ing BOD Removed
Unffltered Effluent
Suspended Solids (mg/1) 346
BOD (mg/1) 14 10 10
BOD Removal (%) 93 95 95
COD (mg/1) 51 53 69
COD Removal (%) 83 84 80
BOD/COD Ratio
Influent 0.66 0.66 0.62
Effluent 0.27 0.19 0 13
Reactor 0.46 0.42 0.35
-------
223
Beginning
of clear
effluent and
sludge
interface
Clear
Effluent
Particles
rising above
sludge blanket
Sludge circulation
pattern
FIGURE 4 Sludge action in upflow clarifier settling tubes
C and above, the SVI increased to values of 260 and greater and the sludge took on a fluffy
snowflake appearance. The rolling action of the sludge in the tubes stopped and the sludge height
began to rise, eventually spilling out with the effluent. Cutting the effluent flow rates back to less
than 0.2 gpm/ft2 resulted in lowering the DO in the effluent tubes to zero, which further
complicated the problem. The studies indicate that some means for backflushing tube-settler
controlled upflow clarifiers must be provided if mixed liquor concentrations greater than 2,000
mg/l are to be achieved with reliable operation.
The 12.45 gal. reactor was operated for a period of time with a very low continuous overflow rate
and then increased to an average rate of 0.5 gpm/ft2 with an alternating on-off cycle. In other
words, with the on 1/2 hour - off 1/2 hour cycle, the actual flow was 1 gpm/ft2 for 1/2 hour. The
SVI again ranged above 200 with very consistent solids removal. The effluent solids concentrations
were very low for the whole range of studies. The longer on times for the on-off cycle (2-1/2 hours
on as opposed to 1/2 hour) did indicate that longer cycles may result in higher effluent solids
concentration. Summaries of the results obtained at various temperatures and overflow rates are
presented in Tables 6 and 7, and Figures 5 and 6. Adding a backwash cycle provided a definite
advantage in that it prevented a bulky sludge from becoming stagnant in the tubes.
Indications are that sludge bulking probably is a general problem in the activated sludge process at
colder operating temperatures and special precautions in design will be necessary to assure
effective solids control. This problem was reported by Ludzack (1965). Bulking sludges have not
been reported in cold temperature oxidation ditch studies (Anonymous, 1965; Grube and Murphy,
1968); however, these ditches were operated at much longer detention times (1.6 to 2.3 days),
which may be a factor. Downing (1968) showed that settleability is improved by longer detention
time |>10 hrs) and very short detention time (<5 hrs) when operating an activated sludge plant at
warm temperatures. At any rate, indications are that backwashing, in conjunction with lower
overflow rates, will overcome this problem.
Pilot Plant
In cooperation with the Alaskan Air Command, the Alaska Water Laboratory constructed and
operated a pilot waste treatment facility at Eielson Air Force Base (EAFB). The facility included
-------
224
TABLE 6
8.9 Gallon Reactor Results of Operation with
Varying Effluent Overflow Rates on the Settling Tubes
INFLUENT REACTOR EFFLUENT
Reactor1 Susp. Susp. Overflow Susp.
Temp Solids BOD COD Solids Rate Solids BOD % BOD COD %COD
(° C) (mg/1) (mgA) (mg/1) (mgA) SVI (gpm/ft2) (mg/1) (mg/1) Removal (mgA) Removal
.35
'.7
(.4-.9)
4.2
(2.8-6.4)
3.8
(3.5-4.1)
95
112
94
77
244
253
193
142
292
370
229
283
3973 238
4237 238
4147 -
4067 229
.4
.3
.6
.3
.6
.5
.8
10
8
20
10
14
10
13
12
22
29
17
20
14
20
95
91
89
91
90
90
86
69
71
87
60
70
62
69
76
79
77
74
69
78
76
(1) Values in parenthesis are minimum and maximum for that period.
TABLE 7
12.45 Gallon Reactor Results of Operation with Varying Effluent Overflow
Rates on the Settling Tubes
INFLUENT REACTOR EFFLUENT
Reactor1 Susp. Susp. Overflow2 Susp.
Temp Solids BOD COD Solids Rate Tube Solids BOD % BOD COD % COD
( C) (mg/1) (mg/1) (mg/1) (mg/1) SVI (gpm/ft2) Size (mg/1) (mgA) Removed (mg/1) Removed
2.4 77 177 303 3957 -- .2 2x3.5 2 19 89 35 88
(1.4-3.5) (continuous)
4x3.5 2 19 89 39 87
2.9 86 185 275 4157 214 .3 2x3.5 4 12 94 50 82
(on 1/2 hr
off 1/2 hr) 4x3.5 3 10 95 52 81
4.4 93 223 321 4095 235 .5 2x3.5 4 12 95 55 83
(4-0-4.7) (onl/2hr
off 1/2 hr) 4 x 3.5 5 12 95 69 79
7.8 87 194 313 4504 209 .5 2x3.5 12 20 90 69 78
(6.8-8.4) (on 1/2 hr
off 1/2 hr) 4x3.5 14 23 88 64 80
(1) Values in parenthesis are minimum and maximum for that period
(2) Notes in parenthesis indicate the time cycle of effluent flow through the tubes
-------
225
50
40-
•s
| 30
I
| 20"
_3
10
0
A
MLSS'~4000mg/l
a * 3—t 7
Overflow Rate (gpm/ft*)
FIGURE 5 8.9 gal. reactor sludge heights in effluent tubes vs. effluent overflow rates with
continuous flow through tubes
20-
|
Umnd
f,
V
A O A
O
G V
contnuouc now
Ot rc
A»4*C
SVffT ft"
off 1/2 hr
D-rc
V-TC
Intftnnhtant Row
on Zhr
off llr
^~8«C
MLSS~400pmo/l
.1 .2 .3 4 .5 Jj 7 A
Owrflow RatM (gpm/ft2)
FIGURE 6 8.9 and 12.45 gal. reactors effluent suspended solids vs. overflow rates at various
temperatures
an aerated lagoon and an extended aeration basin. The purpose of the facility was to increase the
knowledge of biological waste treatment at cold temperatures and to develop design criteria.
Eielson Air Force Base is located 22 miles southeast of Fairbanks. The mean annual temperature at
Fairbanks is approximately 25° F with minimum and maximum recorded temperatures of -66° F
and +99° F respectively (Johnson and Hartman, 1969). The area has approximately 150 days
below 0° F.
Originally intended to serve as, a facultative lagoon, the extended aeration unit consisted of an
earthen basin lined with 20 mil polyvinyl chloride film (PVC) and tube settler modules as shown in
Figures 7 and 8. The PVC film at the bottom of the basin was covered with 6 inches of sand and a
concrete pad poured in the center for support of aerators. Aeration and mixing were provided by
-------
226
eight Hydroshear aerators, manufactured by the Chicago Pump Company. Air supply was by a 120
SCFM Sutorbilt blower, manufactured by the Fuller Company.
Solids separation was provided by two tube settler modules. The tube settlers were developed by
Neptune Microfloc Company for use in water treatment. The manufacturer has recently initiated
studies to adapt them for use in activated sludge separation (Hansen, Culp and Stukenberg, 1967).
This type of settler was felt to provide optimum design for submerged operation which was desired
to overcome icing problems. The basin was fed by a Mar low centrigal pump, manufactured by ITT
Marlow Company. Pumping rate was approximately 180 gpm with the feed drawn from a manhole
on the influent line just before entry to the EAFB primary treatment plant. Temperature of the
sewage averages about 20° C with the sewer lines enclosed in a utilidor, which is heated during the
winter months.
The extended aeration facility was built to provide the very simplest operation with a minimum of
environmental protection for evaluation under cold climate conditions. Construction of the
extended aeration facility was completed in December 1968 and the unit placed in operation later
that month. The unit was operated at a 2-day detention time, which corresponded to an average
overflow loading rate on the tube settler of 1.3 gpm/ft2. A problem was encountered with
breakage of pumps due to entrained solids entering the pumping chamber. The feed line also filled
with solid material and plugged. As a result, the basin was not fed for a week, during which time 3
feet of ice formed over the pond and frozen foam built up to 8 feet above the aerator.
Beginning in January 1969 and lasting approximately 6 weeks, a period of extremely cold weather
occurred with ambient air temperatures dropping as low as -60° F. A detention time of 1 day was
maintained during this period with no ice forming. The loading on the tube settlers was
approximately 2.5 gpm/ft2. The gear housing of a compressor was broken and teeth stripped from
the gears while attempting to start it at a low temperature. Apparently, metal contraction had
reduced clearances which caused internal rotating parts to make contact with and break the pump
housing.
During February, the feed pumps were moved inside the Eielson primary treatment plant and feed
taken from the grit chamber. For the remainder of the winter and the following spring, while
operating at a detention time of 2 days, the MLVSS of the system generally did not rise above 500
mg/l.
Inadequate mixing was suspected as the cause of poor performance, and velocity measurements
were made with an ice current meter obtained from the U. S. Geological Survey which measured
the horizontal component only. Velocities were generally lower than the 1.0 ft per second
recommended for complete mixing, except within 2 feet of the surface. The aeration rate was 120
cfm, depth of the basin 11 feet, with approximately 4 horsepower input. Velocities were measured
again at a later date with 300 cfm being delivered and 9 horsepower input with generally the same
results except the surface velocities were higher. The velocities found were not considered low
enough to cause the extremely poor basin performance.
The possibility of excessive turbulence being carried into the tubes was also considered because of
the close proximity of the settler modules to the aerators (2-3 feet). To check the possibility, a
new aerator was fashioned of a short length of 3-inch pipe attached to flexible hose and placed in
the basin approximately 10 feet from the settler modules. The MLSS of the basin increased to
-------
227
Plan Vbw
Maximum Woridng Liqw) Depth
View A-A
FIGURE 7 Extended aeration pilot facility, Eielson AFB (1968 configuration)
Effluent
Header
FIGURE 8 Tube settler module (1968 configuration)
1,000 mg/l during operation of this aerator which did indicate that basin turbulence or entrained
air bubbles was affecting the settler operation.
The basin was then taken out of operation to permit modifications in preparation for the next
winter's operation. The modifications are illustrated in Figures 9 and 10. The system was placed in
operation in December 1969. It was recognized that at a detention time of 24 hours and with low
winter operating temperatures, the hydraulic load on the tube settlers would be too great. An
attempt was made to reduce the hydraulic load on the system while it maintained a BOD load
equivalent to a 24-hour detention time system by supplementing the feed with primary sludge
from the Eielson treatment plant. Basin velocity proved to be restricted around and beneath the
-------
228
/Cttieaao Pump
(ShcorfuHTS
kTS.
Plan View
Elevation
FIGURE 9 Extended aeration pilot facility after modification, Eielson AFB (1968
configuration)
0 •••
••••
•• .
eo»*
00 »•
«•••«
•••°
«••
o o
« •«
o*«
OQOO
o»» . ^
>0»0 / .-
g^/
•t* '
X /
/ '
•Ji
*v
l^ ^
MO
0»*
\\\\\\\\\\\\\\\\\\\\\
FIGURE 10 Tube settler beneath module design with flow beneath the hopper (1969
configuration)
-------
229
separator hoppers because of the low clearance and resistence offered by the settler support. As a
result, a heavy sludge deposit blocked the separator hoppers, accumulated in the tubes and passed
into the effluent.
During a cold period in January 1970, the surface of the basin began to freeze due to tow heat
energy being supplied. The sludge accumulated in the ice, reducing the suspended solids level in
the pond from approximately 2,500 mg/l to less than 200 mg/l. During this period, the mean
ambient temperatures averaged -23° F, with a range of -8
knots to calm and averaged 3 knots.
'° to -35° F. Wind velocity ranged from 10
A block was cut from the ice and a sample taken of the unfrozen sludge beneath the ice. A cross
section is shown in Figure 11. The ice had reached a thickness of 14 inches with a sludge layer of
17 inches beneath the sampling point. The sludge was not moving under the ice and apparently
had attached itself, building up a thicker and thicker layer which eventually froze into the ice
layer.
Sample
location
Aeration Basin
3"
r"
I
Relatively Clear
12,000 mg/l
18,000 mg/l
30,000 mg/l
T
Ice
Layer
\
I4.0OO mg/l
Sludge
Layer
MLSS before ice build-up= 2500mg/l
MLSS after ice build-up = 400mg/l
Hydraulic detention time' 4 days
Average air temperature -23°F
Maximum -8°F
Minimum' -35° F
Avg. wind during freezeup= 3 knots
Maximum1 10 knots
Minimum-' Calm
FIGURE 11 Profile of surface ice and sludge layers of frozen extended aeration basin
-------
230
The long sloping side walls associated with earthen basins present two very important problems in
activated sludge aeration chamber applications. The relatively high surface area-to-volume ratio
will result in high heat energy losses from the system which may be very critical with low
temperature influent. Greater heat losses will promote ice formation which will entrain MLSS
from the system, destroying the effectiveness of the process.
The second problem is the difficulty in obtaining adequate basin velocities at lower depths without
excessively high horsepower for mixing. Even at high aeration rates, the minimum recommended
velocities of 1.0 fps were generally not present in the pilot facility extended aeration basin at
Eielson Air Force Base.
Another effect observed during operation in the second winter was that, with the aerators off
center, a circular flow was induced in the basin in the horizontal plane around the aerators. The
flow was similar to the Coriolis effect and seemed to be promoted by the earthen basin shape of a
large surface area-to-bottom area ratio. This effect will only become a problem in situations in
which flow directions in the basin are important, as in the Eielson AFB pilot facility, where the
circular flow pattern did have an effect in hindering sludge removal from beneath the hoppers.
The cross sectional shape of a basin and the temperature to which it is exposed will, in general,
determine the type of liner which should be provided. Material must be used which will prevent
erosion and scouring by velocities in the basin. Side slopes of less than 1 vertical to 2 horizontal
permit use of flexible liners, whereas vertical sides will require bearing wall construction of
impermeable concrete or wood crib design with an impermeable liner.
Experience with the PVC liner indicates it is not feasible for use in permanent installations for cold
temperature applications. The liner becomes very susceptible to damage at low temperatures
because of brittleness, and ice formation can cause extensive breaks in the lining. Aging and
exposure to sunlight also increase its susceptibility to damage.
Impermeable liners such as low temperature butyl rubber membranes are feasible for use in
earthen basins when the danger of major freezing does not exist. Care must be taken to insure that
the liner is resistant to hydrocarbons which may be present in the sewage as softening or
dissolution may result.
Concrete provides a reliable material for cold temperature application. However, construction is
expensive in Alaska and particularly so in remote areas. Examples of the successful application of
cheaper methods of concrete construction are the College Utilities oxidation ditch in Fairbanks,
Alaska, and the oxidation ditch at Glenwood, Minnesota (Anonymous, 1965). Concrete block was
used for the construction of vertical sides for the College Utilities ditch. Concrete silo staves were
originally used for the sloping sidewall construction of the Glenwood ditch but were not sealed
and soil behind the staves washed out. The problem was successfully alleviated by placing steel
mesh and 4 inches of concrete grout over the staves to provide a smoother waterproof lining.
CONCLUSIONS
The feasibility of the extended aeration activated sludge process as a relatively economical and
effective means of secondary waste treatment has been demonstrated in the laboratory and in the
field. The process requires more consistent operation and maintenance than aerated lagoons and
this is a disadvantage where costs are high and skilled operators are extremely scarce.
-------
231
The utilization of exposed aeration chambers for the extended aeration process is feasible. Earthen
basins are also feasible for use where economic and construction conditions warrant. When
utilizing exposed basins, heat loss effects must be evaluated in conjunction with detention time
determinations to avoid potential freezing problems. Solids entrainment in ice can cause failure of
an activated sludge process.
Environmental protection in varying degrees should be provided for the remaining equipment,
such as heated enclosures for pumps and flow measurement devices. Housing must be provided for
secondary sedimentation basins and should include a minimum of an unheated structure with
panels which can be removed for warm weather operation.
Effective solids separation is the key to successful operation of extended aeration facilities and is
dependent on both the biological and physical aspects of the system. It has been demonstrated
that a sludge can be developed which will perform very efficiently at temperatures less than 1° C.
A turbid effluent will result at cold temperatures with an unacclimated sludge or loading rates that
are too high. Under these conditions, a less stabilized sludge develops with a corresponding relative
decrease in numbers of stalked ciliates and an increase of dispersed bacteria which appears to
contribute to turbidity (McKinney and Gram, 1956).
A bulking sludge may develop at cold operating temperatures. This type of sludge can lead to
separation problems but will provide a very clear effluent at temperatures ranging down to less
than1°C.
Properly designed tube settlers will provide effective cold (0 to 4° C) temperature solid separation.
This is true for sludges with SVI's ranging up to 250. A backwash cycle should be provided for
reliable operation and is mandatory for operation with high MLSS concentration (4,000 mg/l) and
bulking sludges. Some effort should be directed toward developing upflow clarifier configurations
for cold temperature application since the method has advantages (Reed and Murphy, 1969). The
tube settler does provide an upflow clarifier type action in high MLSS activated sludge solid
separation applications. Providing consistent solids separation with tube settlers at warmer
temperatures (greater than 4° C) appears to be the most demanding and yet insufficiently defined
area of need in their application.
Cold climate sludge wasting and disposal for the extended aeration process must be given
consideration for the following reasons:
1. Excess solids production increases with decreasing temperature.
2. Shorter detention times to prevent freezing will also increase solids production at a given
MLSS level.
3. Auto-induced sludge wasting may be expected to be more severe, placing greater
potential stress on the receiving water.
4. Retarded assimilative capabilities of the receiving water at cold temperatures.
-------
232
RECOMMENDATIONS
Facility Design
The following design recommendations are based on laboratory studies, experience with the
Eielson Air Force Base pilot facility, and experience reported by others:
(1) Exposed aeration basins should be considered for reducing construction costs of waste
treatment facilities. Raw sewage temperatures and heat loss effects must be considered to prevent
freezing which can cause process failure by entrainment of solids from the system.
(2) Housing should be provided for pretreatment units such as bar racks, pumps and flow
measuring equipment.
(3) Some minimum protection should be provided for aeration equipment such as strip heaters or
minimum heat enclosures for compressors and untreated housing for oxidation ditch rotors.
(4) Housing should be provided for secondary sedimentation basins. Minimum housing would
include a structure with panels which may be removed for warm weather operation.
(5) Where economic and construction considerations warrant, earthen basin designs should be
considered for aeration chamber construction. Otherwise, sidewalls that are vertical or nearly so
should be utilized to promote better mixing.
(6) Submerged settling units should be situated in the center of basins with low sidewall
construction with aeration on at least two sides to promote adequate mixing. Several questions
require answers before submerged settling units are practical.
(7) When basins with low sidewall slope construction are utilized without submerged settling
units, the aeration devices should be clustered in the center of the basin for best mixing.
(8) Flexible membranes should not be used where the danger of heavy icing exists.
(9) Concrete block and concrete grout should be considered as economical liner materials where
the design permits.
(10) Tube settlers with backwashing of tubes should be considered. However, more information
is necessary before their reliability can be ascertained. Both routine and emergency maintenance
must be carefully evaluated prior to their use.
Process Design
The following preliminary recommendations for low temperature extended aeration systems are
based on laboratory studies and experience reported by other investigators. Attempts will be made
to verify these findings on a pilot plant scale.
(1) Organic loadings should be maintained below 0.20 Ib. BOD/lb. MLVSS-Day.
-------
233
(2) Provision should be made for sludge wasting of 0.5 Ib. MLSS/lb. BOD removed, particularly
at shorter detention times such as a 12-hour detention time system.
(3) Tube settler overflow rates should be held below 0.5 gpm/ft2 with high MLSS concentrations
{> 4.000 mg/l).
(4) Sludge wasting and disposal facilities or a polishing lagoon for effluent discharge should be
provided where heavy discharges of suspended solids may place excessive stress on receiving
waters.
Research and Development Needs
The following list of suggested research and development needs is not intended to be all inclusive
but includes areas which have come to the attention of the authors through laboratory and pilot
plant experience and a review of experience reported by other investigators:
(1) Sludge bulking conditions at lower temperatures (8° C and below) must be defined so the
condition can be predicted in actual application.
(2) Further develop low temperature biokinetic parameters at detention times ranging from 4 to
36 hours with varying MLSS levels.
(3) Further develop low temperature tube settler design criteria and backwashing techniques at
various MLSS levels.
(4) Investigate upf low clarifier designs for low temperature application.
(5) Investigate methods of sludge digestion and disposal under low temperature conditions;
particularly the use of the freeze-thaw cycle as an aid to promoting better drainability.
(6) Develop reliable methods for positive recirculation of settled solids from submerged settling
units.
(7) Further investigate criteria for predicting heat loss from exposed basins.
(8) Continue evaluation of cold temperature biokinetic design parameters on pilot plants and
existing facilities.
(9) Develop design and operation criteria for low temperature horizontal flow clarifiers.
(10) Investigate power requirements and mixing characteristics of various earthen basin
configurations.
Further investigate the effects of heavy ice cover on solids entrapment in aeration basins,
particularly where flow patterns are parallel to the surface as in the oxidation ditch.
-------
234
REFERENCES
Anonymous (1961) Effect of water temperature on stream reaeration, Thirty-Fjrst Progress
Report, Committee on Sanitary Engineering Research, J. of San. Eng. Div., Proc. Amer. Soc.
of Civil Engineers. 87, No. SAG.
Anonymous (1965) Report on operation of oxidation ditch sewage treatment plant, Glenwood,
Minnesota, Dept. of Health, Div. of Environ. Health, Section of Water Poll. Cont.
Ayres, J. C. (1962) Temperature and moisture requirements, Low Temperature Microbiology
Symposium Proceedings (Camden), Campbell Soup Company.
Benedict, A. H. (1968) Organic loading and temperature in bio-oxidation, Ph.D. Thesis, U. of
Washington.
Black, S. A. (1968) How to evaluate aeration devices. Water and Pollution Control, 106, No. 10.
Culp, G. (1969) A better settling basin, The American City.
Downing. A. L. (1968) Factors to be considered in the design of activated sludge plants. Advances
in Water Quality Improvement, U. of Texas Press, pp. 190-202.
Eckenfelder, W., Jr. (1967) Theory of biological treatment of trade wastes, J. Water Poll. Cont.
Fed., 39, No. 2.
Eckenfelder, W. W. and O'Connor. D. J. (1961) Biological Waste Treatment, Pergamon Press, Inc.,
Long Island, New York.
Frey, P. J. (1969) Significance of winter dissolved oxygen in Alaska, presented Alaska Water
Management Association Annual Meeting.
FWPCA Methods for Chemical Analysis of Water and Wastes, Fed. Water Poll. Cont. Admin., Div.
of Water Quality Research, Analytical Quality Cont. Lab., Cincinnati, Ohio.
Gordon, R. C. (1970) Unpublished data, Alaska Water Lab., College, Alaska.
Grube, G. A. and Murphy, R. S. (1968) Oxidation ditch works well in sub-arctic climate. Water
and Sewage Works, 116, No. 7.
Gustaffson, B. and Westberg, N. (1965) Experiment with treatment of sewage from the town of
Kiruna by the activated sludge method. Royal Inst. of Tech., Stockholm, Sweden, Inst. of
Water Supply and Sewage Tech., Inst. of Water Chem., 65, No. 4.
Halvorson, H. O.. Wolf, J. and Sunevasan, V. L. (1962) Initiation of growth at low temperatures,
Low Temperature Microbiology Symposium Proceedings (Camden), Campbell Soup Co.
Hansen, S. P. and Culp, G. L. (1967) Applying shallow depth sedimentation theory, J. Amer.
Water Works Assoc., 59, No. 9.
-------
235
Hansen, S. P., Gulp. G. L. and Stukenberg, J. R. (1967) Practical application of idealized
sedimentation theory. Presented 1967 Water Poll. Cont. Fed. Conf., New York City.
Heukelekian, H. and Wiesburg, D. (1965) Bound water and activated sludge bulking. Sewage and
Industrial Wastes, 28, No. 4, p. 558.
Hunter, T. V., Genetelli, E. J. and Gilwood, M. E. (1966) Temperature and retention time
relationships in the activated sludge process, Proc. 21st Indust. Waste Conf., Purdue
University.
Johnson, P. R. and Hartman, C. W. (1969) Environmental Atlas of Alaska, Inst. of Arctic Environ.
Eng., Inst. of Water Resources, U. of Alaska, College, Alaska.
Ludzack, R. J. (1965) Observations on bench scale extended aeration sewage treatment, J. Water
Poll. Cont. Fed.. 37, No. 8.
McKinney, R. E. and Gram, A. (1956) Protozoa and activated sludge, Sewage and Industrial
Wastes, 28, No. 10.
Miller, A. P. (1967) The biochemical basis of psychrophily in microorganisms, Inst. of Water
Resources, U. of Alaska, College, Alaska.
Pasveer, A. (1955) Research on activated sludge, V: rate of biochemical oxidation, Sewage and
Industrial Wastes, 27, No. 7.
Pohl, E. F. (1967) The effect of low temperatures on aerobic waste treatment processes, M.S.
Thesis, U. of Washington, Seattle, Washington.
Pohl, E. F. (1970) Chief, Personal Communications, San. Eng. Sec., District Engineers Off., U.S.
Army Corps of Eng., Anchorage, Alaska.
Reed, S. C. and Murphy, R. S. (1969) Low temperature activated sludge settling, J. of the San.
Eng. Div., Proc. Amer. Soc. of Civil Engineers, 95, No. SA4.
Roguski, E., Personal Communications, Alaska Department of Fish & Game, Fairbanks, Alaska.
Schlecta, A. F. and Hsiung, K-Y (1969) High rate processes in advanced waste water treatment.
Presented 1969 Water Cont. Assoc. Pennsylvania.
Schmidtke, N. W. (1967) Low temperature oxidation ditch field study, Thesis to Dept. of Civil
Engineering, U. of Alberta, Edmonton, Alberta, Canada.
Standard Methods for the Examination of Water and Wastewater, 12th Edition, Amer. Pub. Health
Assoc., New York.
Stewart, M. J. (1964) Activated sludge process variations - the complete spectrum. Water and
Sewage Works, pp. R2-41 - R2 62.
-------
236
Thomas, H. A., Jr. (1950) Report of investigation of sewage treatment in low temperature areas,
for the Sub-Committee Waste Disposal, Comm. on San. Eng. and Environ., Nat. Res. Council.
Wuhrmann, K.'(1956) Factors affecting efficiency and solids production in the activated sludge
process, Biological Treatment of Sewage and Industrial Wastes. B. J. McCabe and W. W.
Eckenfelder (ed), Reinhold Publishing Company, New York, New York.
Zanoni, A.E. (1969) Secondary effluent deoxygenation at different temperatures, J. Water Foil.
Cont. Fed., 41 .No. 4.
-------
CHEMICAL TREATMENT OF MECHANICALLY AND
BIOLOGICALLY TREATED WASTEWATER
Arne Rosendahl
INTRODUCTION
Eutrophication problems are found to be serious in many Norwegian lakes and fjords.
An extensive survey of the pollution in the inner Oslofjord has shown that organic matter
produced by autotrophic growth, mostly due to the discharge of nutrients, is 10 - 12 times the
primary organic load from wastewater. In this stratified, landlocked fjord, with a brackish surface
water, the nutrients found to be most important are phosphorus, nitrogen, and iron. With the large
amounts of nitrogen and iron naturally being brought into the system, phosphorus is at the present
the only controllable component.
The same reasoning is found applicable to most of our eutrophied waters, and numerous treatment
plants for removal of phosphorus from wastewaters will have to be installed in years to come.
Today phosphorus removal is thought to be most economical by chemical coagulation/floccutation
and with sedimentation or flotation for the liquid/solid separation.
Several people have studied the theoretical basis for such chemical processes (Stumm and Morgan,
1962; Stumm, 1962; Pope I, 1966; Henriksen, 1962; and Henriksen, 1963} but there is still a lack
of some essential knowledge. This makes it difficult to make a process design on a purely rational
basis, and the treatment plants now in operation are therefore designed on experience gained
through chemical treatment of drinking water during many years, and through experiments made
in new plants for phosphorus removal (Wuhrmann, 1964).
The great majority of these plants are built as tertiary units, i.e. chemical treatment of biologically
treated waste. At a few plants chemical and biological treatment take place simultaneously in the
same reactor (Thomas, 1962).
In Norway there are only a few biological treatment plants, and we have therefore been looking
for a more direct way of phosphorus removal by omitting the biological unit. During 1967 some
treatment plants were built in Sweden using mechanical/chemical treatment, but systematic
observational data describing the process was limited.
Pilot plant studies were needed giving comparable results between mechanical/chemical and
biological/chemical treatment systems, and results and experience that could be translated to
full-scale treatment plants. In 1968 a study'was started at the Norwegian Institute for Water
Research, and results from V/2 years' program are reported in this paper.
237
-------
238
DESCRIPTION OF THE PILOT PLANTS
Five pilot plants were used for this investigation. They were situated at the site of a biological
treatment plant owned by the city of Oslo, treating the effluent of about 50,000 inhabitants, using
the conventional activated sludge method.
One extended aeration plant, type Hycon 5 A, previously used by this institute to study kinetics
of biochemical treatment, was reconstructed. This plant was divided on four parallels based on
coagulation/flocculation and sludge removal by sedimentation (Fig. 1). A fifth separate plant used
coagulation/flocculation and sludge removal by flotation (Fig. 2).
All five plants were in use from January 1969 until May 4, 1970. On that day a serious fire
stopped the experiments at the four sedimentation plants.
Each plant consisted of:
1. Wastewater intake arrangement
2. Chemical feeding equipment
3. Rapid mixing
4. Flocculation basins
5. Sludge-separation equipment (sedimentation or flotation)
6. pH recorder
7. Sampling equipment
Sedimentation Plants
During this investigation primary treated wastewater was used in two plants and secondary treated
wastewater was used in the two other plants.
During the first investigation period a technical quality of aluminum sulfate, type AVR from
Boliden i Sweden, was used. (This type of alum has about 2% impurities, mainly iron and silicates.
It is produced for chemical treatment of sewage and is at the moment the cheapest alum in
Scandinavia). In the last part of the investigation period sulfuric acid was dosed in addition to
alum, for pH adjustments. Both alum and sulfuric acid were made up to a 5% solution. The
chemicals were added to the wastewater ahead of the flocculation basin. To obtain sufficient
mixing the chemicals were added where the water was quite turbulent.
The flocculation basin had four treatment plant paddles mounted on horizontal parallel axes. They
were driven by the same motor, and ran at the same speed. The speed, however, could be varied
within a wide range. Each flocculation basin was 2.5 m long, 0.6 m wide and 0.8 m deep, and was
divided into three chambers by vertical walls with a 30 cm hole around the paddle axes to prevent
short circuiting. Periphery velocity was selected as 0.4 m/sec. This gives theoretically calculated
-------
239
Paddle
Section B, - 6,
Paddle
Section B, — Bj
Paddle
Section B3 - B3
1 Dosing pump
main chemical
1 Dosing pump
•ck) or coagu-
lation aid
3 Motor for paddles
4 Sludge removal
pump
5 Sampling pump
6 Refrigi rater
for samples
7 Automatic instru-
mentation for
sludge removal
8 pH-instruments
x pH-electrodes
. Sampling points
Chambers for
water distri-
Holding tanks I butionand
for chemicals I quantity
control
Biologically treated influent!
- Mechanically treated influent' Surplus water and
sludge effluent
Holding tanks
for chemicals
Distribution
chambers
H.AN
Flocculation unit
Sedimentation unit
Sampling
Effluent
Section A — A
FIGURE 1 Sedimentation plants
-------
240
1 Influent
2 4 rapid mix chamber
3 Rapid mixer for scum-control
4 First flocculation compartment
5 Second f tocculatkm compartment
6 Paddle with motor A
7 Flotation unit
8 Shut off valve for effluent
9 Effluent
10 Sludge outlet
11 High pressure 1>ump
12 Level regulate*
13 Pressure tank
14 Compressor
Qlm
0,1m
0,7"
PLAN
IS Rotameter
16 Reduction v*lve for dispersion
17 Holding unk* for main chemical
18 Dosing pump tor 17
19 Motor»ndpaddle forchemical
holding tanks
20 Holding tank for acid or
coagulation aid
21 Dosing pump for 20
22 Sampling pump
23 Refrigirator for samples
24 Sampling point
25 pH-inctrument
26 pH-electrode
27 Motor for sludge scraper
28 Sludgescraper
,15m O.7n
SectionA-A
This equipment is placed
on the tame level in
front of the flocculdtion
compartments
FIGURE 2 Flotation plant
-------
241
g-values of 59 sec"1. 55 sec"1 and 46 sec"1 in the first, second and third chambers (Camp. 1943).
The influence of g-value variation at the plant has been studied as a diploma thesis for Civil
Engineering in the autumn 1969.
The sedimentation units each had a length of 2.5 m. the width being 0.6 m. The bottom had steep
slopes to the middle and the deepest point was 2.7 m below the surface. Water entered the basin
from one end, flowing about horizontally through the basin and leaving at the other end over a
V-notch weir. Sludge was withdrawn automatically for 30 seconds every hour.
Equipment for pH recording was installed in autumn 1969, and pH values were recorded on both
mechanically and biologically treated wastewater before adding chemicals, and in the first
flocculation chamber of each plant. Sampling was done continuously during every 24 hour period,
and the bottles placed in a refrigerator.
The hydraulic load was 1.2 m3/hr (5.28 gal/min) to each unit. This load was selected after trial
runs with varying loads. Sludge loss from the sedimentation units was kept at a practical level. This
toad gave a theoretical detention time of 1 hr in the flocculation units, about 1.5 hr detention
time, and theoretical overflow rate of 0.8 m/hr (470 GPD/Sq Ft) in the sedimentation units.
The Flotation Plant
Primary effluent was treated in the fifth plant throughout the whole investigation period. 10%
solution alum as well as sutfuric acid was used.
The chemicals used were mixed with the wastewater in a baffled rapid mixing basin. The plant was
provided with two flocculation chambers coupled in series, with separate paddles on vertical axes.
Each flocculation chamber had an area of 0.7 x 0.7 m and a depth of 1.06 m. Paddles ran with a
peripherical speed of 58.5 cm/sec in the first chamber and 35 cm/sec in the second chamber. This
gave theoretical g-values of 78 sec"1 and 37 sec"1 in the first and the second chamber.
The flotation unit itself was circular with a diameter of 97 cm and a depth of 37.5 cm, made of
concentric cylinders and cones. Flotation was effected by small air bubbles. They were produced
by solute air in water under 5 atm. pressure. The pressure was released just before adding the
flotation medium to the flocculated wastewater.
The sludge floated at the surface while the water flowed downwards and left the unit through four
pipes at the periphery. The four pipes were connected to a larger pipe. The water level in the unit
was controlled at the outlet. About 15% of the effluent was used to make up the dispersion water.
This quantity was manually controlled by a flowmeter.
Sludge withdrawal was automatically" controlled by a timer. Sludge was removed by closing the
outlet valve for approximately 2 minutes every hour. While the water level increased, sludge and
water left the plant over a mechanically cleaned weir.
The flotation plant was designed for 2 m3/hr (8.8 gal/min) loading. This hydraulic load was kept
constant throughout the test period. This gave a total theoretical detention time of 31 minutes in
the flocculation chambers. Theoretical detention time in the flotation chamber was 5.1 minutes
with the overflow rate of 4.3 m/hr (2520 GPD/Sq Ft). The plant had a pH recorder. Composite
-------
242
sampling was done during the complete test. Samples of the raw wastewater entering the
community plant were also analyzed.
ANALYZED PARAMETERS
Several parameters were considered for the evaluation of the quality of the influent, effluent and
kinetics of the processes.
The following parameters were analyzed:
Organic matter: chemical oxygen demand (COD) by the dichromate method
Biochemical oxygen demand, seven days test (BOD7)
Total phosphorus, orthophosphate
Nitrogen: Kjeldahl and the sum of nitrite-nitrate nitrogen.
Turbidity (JTU)
pH, continuously recorded in flocculation chamber, and measured on day-samples in the
laboratory.
Aluminum
TABLE 1
Mechanical/Chemical Treatment
Sludge Separation by Sedimentation
Chemical Dosed: Alum
Water * Chemical dose
quality alum mg/1
R
M
M/C 75
R
M
M/C 100
R
M
M/C 125
R
M
M/C 150
Parameters
No. of day COD
mples
re. value
4
4
4
8
6
5
4
4
4
8
6
5
(K2Cr,07)
mgO/1
186.4
166.6
91.1
158.6
104.3
59.0
186.4
166.6
42.1
158.6
104.0
44.0
BODj
mgO/1
94.5
48.0
35.0
98.0
49.6
18.3
94.5
48.0
10.5
98.0
49.6
17.5
Tot.P
mgP/1
4.65
4.58
3.75
4.45
4.13
1.61
4.65
4.57
0.14
4.45
4.13
0.14
Kjeldahl N
mgN/1
19.2
20.9
17.5
17.3
19.2
13.2
19.2
20.9
17.6
17.3
19.2
14.3
Turb.
JTU
61.0
28.5
15.8
51.8
14.6
10.8
61.0
28.5
1.5
51.9
14.6
0.9
pH
7.30
7.11
7.12
7.56
7.39
6.86
7.30
7.11
6.74
7.56
7.39
6.42
-------
243
Calcium, magnesium and alkalinity were measured on a series of samples to obtain the average
hardness of the treated water.
The process variables selected were chemical dosage, hydraulic loading and flocculator speed
(g-values).
RESEARCH PROGRAM AND RESULTS
The first research series covered doses of 75, 100, 125 and 150 mg/l alum. These doses were added
to biologically treated wastewater and mechanically treated wastewater.
The results are given in Tables 1, 2 and 3, and some results are shown graphically in Figure 3.
During these experiments it was obvious that quality of the effluent varied extremely during the
day. The results seemed to follow the variations of pH which could vary within ± 1.0 pH unit
during the day, with the lowest pH at 6:00 a.m. and the highest value about 1:00 p.m. Turbid and
colored effluent was noticed when the pH was higher than 6.2 - 6.5 in the flocculation chamber. It
looked as if a critical pH value was lower at a lower dose of alum. From this it was decided to try
alum doses of 70, 80, 90 and 100 mg/l and adjust pH to a maximum of 6.5 using sulfuric acid. The
results of these experiments are listed in Table 4 and some results are shown graphically in Figure
4.
TABLE 2
Mechanical/Chemical Treatment
Sludge Removal by Flotation
Chemical Dosed: Alum
Parameters
Water*
quality
R
M
M/C
R
M
M/C
R
M
M/C
R
M
M/C
Chemical dose
alum mg/l
75
100
125
150
No. of day
samples
for ave. value
10
8
10
4
4
4
8
9
9
7
5
7
COD
mg
ro
O/1
231.4
149.0
92.4
137.5
101.4
45.8
209.1
123.9
47.4
194.3
109.5
34.4
BODj
mgO/1
122.0
58.4
25.8
83.5
47.0
14.8
114.0
47.0
10.9
132.8
40.3
8.7
Tot.P
mgP/1
5.04
4.63
2.84
4.43
3.98
0.99
5.88
4.53
0.32
4.68
4.38
0.19
Kjeldahl N
mgN/1
21.5
20.9
19.2
16.9
19.7
15.2
28.1
20.2
19.9
19.5
17.5
15.6
Turb.
JTU
96.1
26.6
20.0
29.9
13.3
5.6
34.0
16.2
1.1
62.4
15.9
1.3
pH
7.23
7.05
7.07
7.54
7.44
7.04
7.50
7.27
7.02
7.46
7.14
6.61
R = Raw wastewater
M = Mechanically treated
B = Biologically treated
C = Chemically treated
-------
244
in —
|
.
\
\
\
\
V ]|
h"-^
-
Mean values:
Raw sewage 203.
Mechanically treated 125.
Biologically treated 53.
~"- <
I J
»_ f
-
75 100
Alum-dose mg/l
125
150
Mean values:
^
o
?.,
r-
n in —
8 I0
m
(
y v
\
v
vj
-
>> —
^^»— ^i
Rawsewa
Mechanic
Biologica
•N^
^-^.
'
1
ge 111.9mgO/t
ally treated 49.4 mg O/l
ly treated 1 0.4 mg 0/1
u
1 -*
1
75 100
Alum-dose mg/l
125
ISO
CL
1
5
S
|
1 1
|
ft
7
^
V
\
\
\
\ \
\
Mean vak
Raw sewa
Mechanic
M
ge 5.35
ally treated 4.44
Biologically treated 3.86
\
"^O! ^X^
p^-^^j
1^-- ^
f
i i
5 100 125 150
Alum-dose mg/l
Symbols:
meal treatment "I Sllia^ removai by sedimentation
imical treatment J
A Mechanical-chemical treatment Sludge removal by flotation
FIGURE 3 Results by dose of alum at sedimentation and flotation units on mechanically and
biologically treated wastewater
-------
245
The pH values given in the tables are average values of daily samples measured in the laboratory.
They are approximately 0.5 pH units higher than the value measured in the flocculation chamber.
The influence of pH will be studied further on the fifth unit (using flotation for sludge removal)
where a full pH control will be carried out.
The values in Tables 1 to 4 are average for several 24 hr-samples. The samples are mostly from a
period of about 14 days. The number of samples from which the average values were calculated
varied from 3 to 10. The series with the lowest numbers were supposed to be redone. The points
plotted in the diagrams are from the tables.
Phosphorus Removal
From January until May 1969 the phosphorus content of the raw waste averaged 5.35 mg/l P.
Mechanically treated water from the community plant had in the same period an average of 4.44
mg/l P and the biologically treated water had 3.86 mg/l P.
The curves in Figure 3 show an increase in phosphorus removal with an increasing dose of 75 to
125 mg/l alum for both mechanically as well as biologically treated waste water. From 125 to 150
mg/l there is only a slight increase in the quality of the effluent regarding phosphorus removal.
At the dose of 75 and 100 mg/l with alum there is a significant difference in quality of the effluent
between biologically and mechanically pretreated waste used. Thus, effluent with about 1 mg/l P
can be achieved when treating biologically pretreated wastewater and 3 - 3.5 mg/l P when treating
Water*
quality
R
B
B/C
R
B
B/C
R
B
B/C
R
B
B/C
TABLE 3
Biological/Chemical Treatment
Sludge Removal by Sedimentation
Chemical Dosed: Alum
Chemical dose
alum mg/l
_
.
75
.
-
100
.
.
125
.
-
150
No. of day
samples
for ave. value
4
4
4
7
5
4
4
4
4
8
6
6
COD
(K2CT207)
mgO/1
186.4
59.1
49.3
154.2
49.3
38.8
186.4
59.1
33.3
158.6
50.6
34.1
BOD-T
mgO/1
94.5
12.0
11.5
97.0
11.6
8.5
94.5
12.0
3.5
98.0
10.0
4.7
Parameters
Tot. P
mgP/1
4.65
4.00
1.08
4.45
3.68
0.70
4.65
4.00
0.07
4.45
3.77
0.09
Kjeldahl N
mgN/1
19.2
17.1
15.9
17.3
14.5
14.0
19.2
17.1
15.1
17.3
14.9
14.1
Turb.
JTU
61.0
4.3
2.7
47.5
1.7
1.4
61.0
4.3
0.5
51.8
2.0
0.3
PH
7.30
7.31
6.98
7.59
7.31
6.76
7.30
7.31
6.93
7.56
7.29
6.58
-------
246
mechanically pretreated wastewater using a chemical dose of 75 mg/l alum. At a dose of 100 mg/l
alum about 0.5 mg/l 0 and 1-1.5 mg/l P can be obtained using chemical treatment respectively.
At the dose of 125 mg/l alum the results vary from 0.07 to 0.32 mg/l P, and at the dose of 150
mg/l from 0.09 to 0.19 mg/l P. Again the best results using chemical treatment were achieved with
biologically pretreated wastewater. At these doses the difference between the quality of the
wastewater, pretreated mechanically or biologically, is not significant.
Using acid to lower the pH caused a marked difference in the quality of the effluent from 70 to 90
mg/l alum. From 90 to 100 mg/l alum there was only a slight increase in the quality.
It seems as if the same results can be obtained using a dose of 90-100 mg/l alum and pH control
compared to a dose of 125 mg/l alum without pH control regarding phosphorus removal.
Orthophosphate removal followed the trend of total phosphorus removal very closely.
Removal of Organic Matter
BOD7 and COD of the influent and effluent from the chemical treatment plants, as well as the raw
water entering the community plant, were measured. The curves and tables show that BOD and
COD removals follow each other very closely.
TABLE 4
Mechanical/Chemical Treatment
Sludge Removal by Flotation
Chemicals Dosed: Alum and Acid
Chemical dose
Parameters
Water *
quality
R
M
M/C
R
M
M/C
R
M
M/C
R
M
M/C
Alum
mg/l
.
-
70
.
-
80
.
-
90
.
-
100
H2S04
mg/l
-
-
78
.
-
54
.
-
42
.
-
30
No. of day
samples
for ave . value
8
8
8
5
5
5
4
4
4
3
3
3
COD
(K2Cr207)
mgO/1
172.1
199.0
61.0
201.1
193.3
60.0
155.0
156.2
39.4
178.6
132.5
31.6
BODj
mgO/1
98.4
132.3
17.6
87.0
104.5
18.0
91.8
125.5
15.5
102.0
90.5
7.3
Tot.P
mgP/1
4.37
5.61
2.41
5.04
5.76
1.74
4.75
4.68
0.56
4.90
5.00
0.25
Kjeldahl N
mgN/1
18.5
22.9
17.9
22.6
28.0
21.1
20.8
19.8
16.7
21.8
22.5
16.2
Turb.
JTU
30.1
45.9
5.3
48.4
109.0
6.7
54.0
40.1
2.5
31.7
22.2
0.7
PH
7.66
8.03
5.11
7.57
7.58
6.48
7.29
7.28
6.72
7.24
7.32
6.72
R = Raw wastewater
M = Mechanically treated
B = Biologically treated
C = Chemically treated
-------
247
o
o
n
r*
* in
Q .
\
N
\
\
N
x
"N
X
\
X.
•-^
N^
Raw
Mec
l_
70 80 90
Alum dose mg/l
ige 1 76.9 mg 0/1
Mechanically treated 178.0 mg 0/1
Mean values:
Raw sewage 94.1mgO/l
Mechanically treated 118.4 mg O/l
70 BO 90 100
Alum dose mg/l
Q.
P 3
4
§
'
\
\
\ s.
\
\
N
V
V
\
\
N
X
•^
^
s
Ml
R«
M
h^
-
age 4.47 mg P/l
Mechanically treated 5.37 mg P/l
70 80 90 100
Alum dose mg/l
O Mechanical-chemical treatment on flotation unit with dose of acid
Mechanical-chemical treatment on flotation unit without do» of acid
FIGURE 4 Results by dose of alum and acid at flotation unit on mechanically treated
wastewater
-------
248
The difference in the removal is larger between mechanical/chemical treatment and
biological/chemical treatment for organic matter compared to phosphorus removal. This is also
true when a high alum dose is used.
A dose of 75 mg/l alum gives only a slightly lower BOD and COD with chemical treatment. An
increase in dose up to 125 mg/l alum gives better removal, especially on mechanically pretreated
waste. An increase in chemical dose from 125 to 150 mg/l alum gives only a small improvement in
organic matter.
It is most important that mechanical/chemical treatment using a dose of 125 to 150 mg/l alum
gives an effluent regarding organic matter which has the same quality as biologically treated
wastewater. From the curves and tables it is shown that biological treatment gives a COD
reduction from 203 to 53 mg/l, and BOD7 reduction from 112 to 10 mg/l. With a dose of 125
mg/l alum using mechanically treated water, an effluent of 45 mg COD and 11 mg/l BOD was
achieved.
The results from a combined dose of 100 mg/l alum and pH adjustments are somewhat better than
using 125 mg/l alum without pH adjustments. The curve is steeper in the range of 90 - 100 mg/l
alum with acid than it is in the range of 125 -150 mg/l alum without acid. It seems therefore that
a higher dose of alum could give better results regarding organic matter when pH adjustments are
made.
Other Components
Removal of turbidity followed phosphorus removal very closely. Aluminum residual was measured
in the effluent and was thought to be a good parameter to control the process, but experience
showed that turbidity and pH gave better information on the conditions of the plant, and were
easier to observe. Turbidity and pH seemed to be the parameters most likely to indicate the
efficiency of the process.
Nitrogen has been analyzed as Kjeldahl-nitrogen and as a sum of nitrite-nitrate nitrogens.
Reduction of Kjeldahl-nitrogen was moderate by chemical treatment when using alum as a
coagulant both with or without control of pH. There were only slightly better results when the
chemical dose was increased. The amount of nitrogen was reduced from 20 - 15 mg/l N in the most
effective experiments.
Nitrite-nitrate has always been below 1 mg/l N in the raw mechanically and biologically treated
water. By the chemical treatment there has always been a tendency for higher values of
nitrite-nitrate in the effluent compared to the influent for mechanically and biologically pretreated
wastewater. However, the values have always been lower than 1 mg/l N except in two cases.
Nitrite-nitrate always makes up less than 5% of the total nitrogen.
Average value of hardness for the wastewater treated was 1.7 dH. The experiments using alum and
acid show higher values of most parameters for mechanically treated water as for the raw water.
The reason for this has not yet been found. It may have a connection with the accumulation of
suspended solids in the return surplus sludge from the biological units to the mechanical units.
-------
249
TREATMENT COST AND PRACTICAL EXPERIENCE
The type of alum used so far is delivered at a price of 350.-N.kr/ton in Oslo. The price for sulfuric
acid is 225.-N.kr/ton in 95% concentration. The prices are:
125 mg/l alum 4.37 0re/m3
100 mg/l alum + 30 mg/l H2SO4 4.21 0re/m3
(1 N.kr = 100N.0re = $0.14)
The difference in price when using one or two chemicals is small. The prices will differ for
different places in Norway, and other countries may have different prices. Treatment efficiency
and economical aspects should be evaluated in each case.
Sludge production was measured, but is not presented in this paper. It is obvious, however, that
sludge production decreases with decreasing doses of alum. This may be a deciding factor for
choosing two chemicals rather than one.
It will probably be unusual to use two chemicals at smaller treatment plants. (In Norway there will
be a large number of treatment plants serving less than 5,000 persons). The lower cost and smaller
sludge production by using two chemicals will not outweigh the extra operational difficulties.
At larger treatment plants, however, reduction in operational costs is more important. It is more
economical to invest more in equipment and automatic control of chemical dosing to reduce the
cost of operation. Reduction in sludge production will also be more important at large plants. At
such plants one will get full advantage through the combined dose of alum and acid, if prices are
such that this combination gives the lowest running cost.
Flotation is not suggested for small scale treatment for economical and operational reasons. This
process has more mechanical equipment and requires highly skilled personnel. Effluent to be used
as dispersion water should have good quality to avoid clogging.
For larger plants these reservations should be easily overcome. Using flotation decreases the
construction cost even though the process needs more mechanical equipment.
The type of alum used for the experiments contains some very fine insoluble impurities. These
have a tendency to settle in pipes before they are mixed with the wastewater, thus clogging valves
and pumps. They may also cause wear on pumps. The company Boliden, which is delivering this
alum, has made suitable equipment for dosing the chemical.
It |s assumed that chemical treatment is less affected by low temperatures than biological
treatment. Therefore the effect of temperature was not considered during these experiments. The
testing was run under the climatic conditions in Oslo which are not extreme. Outdoor
temperatures may be 20 - 25° C below zero for about one month. The wastewater temperature is
decreased to about 5° C.
Construction costs for biological and chemical treatment is evaluated by Dr. Wuhrmann (1967) to
be about 30% higher than conventional biological treatment by the activated sludge method. The
costs for mechanical/chemical treatment plants were estimated to be about 20% less than the costs
-------
250
for conventional biological treatment plants. Such a consideration will be valid for treatment
plants larger than about 5,000 persons. Costs of smaller chemical treatment plants can be higher
than biological treatment plants.
With the results from the experiments regarding removal of phosphorus and organic matter, it
seems that mechanical/chemical treatment should be considered first where eutrophication is the
greatest problem in the receiving waters and phosphorus is the most important nutrient to be
removed from the wastewater.
FUTURE INVESTIGATIONS
We had an extensive program for future investigations using the five plants. It was interrupted by
the fire in May, when the four sedimentation plants were totally damaged. There is a great need
for further investigations, and the work will be continued.
The experiments using the flotation plant will be continued with Norwegian alum. pH equipment
will be enlarged to cover automatic dosing of chemicals through pH control. Studies will then be
continued to find the most suitable and economic operational conditions using alum and acid.
Further, it is of great interest to find practical methods for using ferro-sulfate as a coagulant. This
is an industrial waste product and will probably be the cheapest coagulant in Norway. From
laboratory studies (Henriksen, 1962 and 1963} it seems that it should be an easy and inexpensive
way to oxidize the iron from di-valent to tri-valent before or during the coagulation process.
We are also interested in studying coagulation using lime and stripping of ammonia by aeration at
high pH.
All investigations will be preferably carried out simultaneously on mechanically and biologically
treated wastewater. We are also interested in studying direct chemical treatment on raw
wastewater.
REFERENCES
Camp, S. (1943) Velocity Gradients and Internal Work in Fluid Motion, Boston Soc. Civ. Eng., p.
219.
Henriksen, A. (1962) Laboratory studies on the removal of phosphates from sewage by the
coagulation process, Schweizerische Zeitschrift fur Hydrologie, Vol. 24, p. 253.
Henriksen, A. (1963) Laboratory studies on the removal of phosphates from sewage by the
coagulation process, Part 2, Schweizerische Zeitschrif t fur Hydrologie, Vol. 25, p. 380.
Popel, J. (1966) Die Elimination von Phosphaten, Kommisionsverlag R. Oldenburg, Munchen.
Stumm, W. and Morgan, J. (1962) Chemical aspects of coagulation, J. Amer. Wat. Works Assoc.,
Vol. 54, p. 971.
Stumm, W. (1962) Discussion to paper by Rolich, G. A.: Methods for the removal of phosphorus
-------
251
and nitrogen from sewage plant effluents. Advances in Water pollution research, Proc.
Internal. Conf. London, Vol. 2, p. 216.
Thomas, E. A. (1962) Verfahren zur Entfernung vor Phosphaten aus Abwassern,Schweiz, Patent
361543.
Wuhrmann, K. (1964) Stickstoff-und Phosphorelimination, Ergebnisse von Versuchen im
Technischen Masstab, Schweizerische Zeitschrift fur Hydrologie, Vol. 26, p. 520.
Wuhrmann, K. (1967) Probleme der dritten Reinigungsstufe von Abwassern, Federation
Europaischer Gewsser schutz (PEG) Informationsblatt Nr. 14.
-------
BIOLOGICAL AND CHEMICAL WASTE TREATMENT EXPERIMENTS
IN FAR NORTHERN SWEDEN
Peter Balmer
INTRODUCTION
,,wn of Kiruna is situated at 68° north latitude (Fig. 1) and is hence the northernmost town
in Sweeten In the winter, temperature extremes of 35 C ( 3CT F) occur
The existing primary treatment plant was overloaded and the town planned to replace it with a
new plant with primary and activated sludge treatment
The know how of design of an activated sludge plant for tre<
limited at this time (1962 1963) In a literature rev.ev
Ijijoutory data clearly showed that the metabolizing ai
i.-
-------
253
EXPERIMENTS WITH BIOLOGICAL TREATMENT
Activated Sludge Experiments
Apparatus
The pilot plant consisted of three aeration tanks mounted in series wtih working volumes of 2.0
m3 (71 cu ft), 2.0 m3 (71 cu ft), and 1.7 m3 (61 cu ft). The settling tank was conical with a
volume of 2.7 m3 (96 cu ft) and a surface area of 2.5 m2 (270 sq ft). The tanks were provided
with inserted walls so that back-mixing was minimized. Air was supplied through perforated plastic
piping on the bottoms of the tanks. Sampling (24-hour composites) was done with automatic
samplers.
Sewage
The experimental unit was supplied with presettled sewage from the existing plant in Kiruna. The
Kiruna sewage is almost entirely of domestic origin. The strength of the sewage is, however, quite
low.
Experiments
During June, 1963 to May, 1964, 6 runs were made. The details are extensively described in
duplicated reports (Balmer, Berglund and Granstrand, 1964; Balmer, Berglund and Widell, 1964)
(in Swedish) and summarized by Gustavsson and Westberg (1965).
The operating conditions during the runs are given in Table 1, and detention times and calculated
load factors are shown in Table 2. In all runs except run 3 the unit was run as a conventional
activated sludge process. In run 3 the unit was run according to the contact stabilization principle
with step addition of return sludge (Balmer, Berglund and Enebo, 1967). Flow-sheets for the runs
are given in Figure 2.
TABLE 1
Mean Operating Conditions
Period
1
2
3*
,4
5
6
Length
of run
days
33
34
17
19
11
24
Temp.
9.4
7.8
6.3
5.8
3.7
5.4
Sewage
flow Recirculation Air flow
m3/hr ratio m /hr
1.47
1.36
1.21
0.96
1.19
1.80
0.42
0.42
0.45
0.42
0.34
0.33
14
28
31
24
26
21
Suspended solids
in aerator
ppm
3,000
3,600
3,300/4,200
3,600
3,200
2,300
Oxygen concentration
tank 1 tank 2 tank 3
ppm ppm ppm
0.4
0.7
0.0
0.9
4.4
2.3
0.8
1.0
0.0
1.1
4.0
2.3
1.0
1.4
0.4
2.7
6.9
2.3
*Step addition of return sludge. Suspended solids concentration in return sludge aeration tank
12,400 ppm.
-------
254
PERIOD 1-2, 4-6
PERIOD 3
AERATION TANK
EFFLUENT
t * INFLUENT
R * RETURN SLUDGE
S - SETTLING TO*
SA* SLUDGE AERATION TANK
FIGURE 2 Flow sheet of experimental runs (June 1963-May 1964)
Results
Mean results from the 6 runs are presented in Table 3. In period 1 steady state conditions were
probably not reached.
The oxygen content in the aeration tanks was very low throughout periods 1 - 3, indicating that
oxygenation capacity of the equipment was the limiting factor. In periods 4 - 6 the oxygen
content in the tanks was higher although the oxygen supply was lower compared to periods 2 and
3. It is therefore probable that bacterial activity was the rate determining factor in these periods.
The mean temperature difference between periods 3 and 4 is small and the difference in bacterial
activity is probably not as large as the differences in oxygen concentration in the aeration tanks
may indicate. The low oxygen concentration during period 3 (contact stabilization) is explained
by the large amount of suspended solids in the reaeration tank (tank 1} and the rather high load on
the two aeration tanks.
The conclusions of the activated sludge experiments in pilot plant scale were:
Biological treatment is possible at temperatures as low as 3° - 4° C (37° - 39° F). The activity
of the activated sludge is seriously affected by low temperatures. Long aeration periods are
necessary. If a low BOD sewage is treated, a period of at least 4 hours is required to reach an
effluent BOD of 20 - 25 ppm.
The contact stabilization technique is attractive as it gives a larger amount of sludge actively
metabolizing in a given aeration volume.
As biological treatment now was proved feasible, the town planned to build a treatment plant
-------
255
Detention time
aeration
Period hr
1 3.9
2 4.2
3 3.1*
4 5.9
5 4.8
6 3.2
TABLE 2
Load
Detention time
settling
hr
1.8
2.0
2.2
2.8
2.3
1.5
Factors
Surface load
settling
m /m* , hr
0.59
0.54
0.48
0.38
0.48
0.72
BOD-load
kgBOD/m3,d
0.34
0.46
0.41
0.40
0.40
0.41
Sludge load
kg BOD/kg
sludge, d
0.11
0.13
0.06
0.11
0.13
0.25
*Detention time in aeration tanks only. Detention time in return sludge aeration tank was 3.7 hr.
consisting of presettling and secondary treatment with activated sludge. The primary treatment
unit was first erected and was put into operations in 1967.
EXPERIMENTS WITH CHEMICAL TREATMENT
In the years since the activated sludge experiments, there has been a complete turnover in the
opinion on the value of biological treatment. Biological treatment is now considered of very
limited value as the initial BOO load caused by domestic effluents in the receiving waters is
considered low in relation to the secondary pollution caused by nutrients, mainly phosphorous
compounds.
It was then discussed whether the planned addition of aeration tanks and secondary settling tanks
could be replaced advantageously with a coarse presettling basin and a f locculation basin placed in
front of the existing settling tanks, as illustrated in Figure 3.
TABLE 3
Analytical Results - Activated Sludge Treatment
INFLUENT
EFFLUENT
Period
1
2
3
4
5
6
BOD 5
ppm
55
81
80
83
82
76
Suspended
solids
ppm
46
84
56
54
54
51
Setteable
solids
ml/1
1.7
1.0
0.7
1.0
1.1
0.4
BODS
ppm
13
15
10
20
20
24
Suspended
solids
ppm
29
14
16
18
21
7
Setteable
solids
ml/1
0.5
0.1
0.1
0.1
0.1
0.1
No. of
composite
samples
10
12
4
9
6
9
-------
256
ACTTVATEO SLUDGE TREATMENT
CHEUCAL TREATMENT
A-AERATION BASIN
F'FLOCCULATKM BASIN
S* SETTLING BASIN
DOUBLE UNE • EXISTING UNIT
FIGURE 3 Two ways to enlarge a primary treatment plant
As a preliminary economic analysis also indicated that chemical treatment could be cheaper, this
seemed to be an attractive alternative. In order to determine the treatment efficiency of a chemical
treatment process, the town had to carry out pilot plant experiments.
Experiments with Flocculation and Settling
Apparatus
The pilot plant equipment (Fig. 4) consisted of a flocculation tank and a settling tank. The
flocculation tank was divided into 4 compartments, each with a working volume of 0.23 m3 (8 cu
ft). The compartments were provided with paddle-type mixers with peripheral speeds of 0.5. 0.35,
0.25, 0.15 m/s (1.6. 1.2. 0.8, 0.5 ft/s) respectively. The settling tank was rectangular with a
volume of 2.0 m3 (70 cu ft) and a surface of 2.0 m2 (220 sq ft).
The flocculated sewage was transferred from the flocculation tank to the settling tank through a
siphon with a diameter of 10 cm (4 in). Sampling (24-hour composites) was done with automatic
samplers.
The equipment proved suitable for its purpose although the peripheral speeds of the mixers in the
2 last compartments was somewhat too fast. The transfer from the flocculation tank to the settling
tank also destroyed some floes.
Sewage
The pilot plant equipment was fed with domestic sewage that was pumped from the inlet zone of
one of the primary settling basins in the new treatment plant. In this way a coarse presettling was
-------
257
FIGURE 4 Pilot plant for chemical treatment
simulated. The sewage was almost entirely of domestic origin and the temperature of the sewage
was about 3° C (37° F).
Experiments
In 4 experiments different amounts of technical grade alum (8.1% Al) were added to the sewage.
In one of the runs, addition of an anionic polymer (Dow Purifloc A-23) was tried.
The flow through the apparatus was 0.96 m3/hr (4.2 gpm) which means a detention time in the
flocculation tank of about 1 hr, and of 2 hr in the settling tank.
Results
Mean results from the test runs are given in Table 4.
Experiments with Simplified Chemical Treatment
The Treatment Plant
The sewage treatment plant in Kiruna consists of a coarse bar screen, an aerated grit chamber, a
comminutor and three parallel settling basins. Each settling basin has a width of 6 m (20 ft) and a
length of 47 m (148 ft) and has a total volume of 640 m3 (22,800 cu ft). During the test week the
flow through the plant was almost the same every day. Flow and load factors of the settling basins
are given in Figure 5.
-------
258
TABLE 4
Analytical Results - Chemical Treatment
INFLUENT EFFLUENT
FILTERED EFFLUENT
Alum No. of
dosage Flocculation BOD7 COD Phosphorus BOD7 COD Phosphorus BOD7 Phosphorus Composite
ppm
100
120
120*
150
PH
6.6-6.9
6.4-6.9
6.4-6.9
5.2-5.5
ppm
60
50
52
81
ppm
24
25
30
33
ppm
6.5
2.9
4.3
3.9
ppm
41
27
26
33
ppm
18
15
19
19
ppm
2.3
0.62
0.64
0.79
ppm
28
24
20
25
ppm
0.20
0.39
0.35
0.10
samples
4
3
3
3
*1 ppm Dow Purifloc A-23 added. COD was determined with potassium-permanganate oxidation.
FLOW
900
800
TOO
600
500
400
300
200
CO
— FLOW
— SURFACE LOAD
— DETENTION TIME
(2
DETENTION TIME, h
SURFACE LOAD m /m, h
2
FIGURE 5 Wastewater flow and detention time and surface load on settling basins
-------
259
TABLE 5
Operating Conditions - Simplified Chemical Treatment
Date Time Alum dosage Flocculation-pH
ppm
12/1 1100-2200 141 6.0-6.4
13/1 1300-2200 142 5.8-6.1
14/1 0900-1700 122 6.6-6.8
15/1 0800-1600 117 6.0-6.4
15-16/1 2200-1300 90 5.7-6.2
17/1 0500-1000 100 6.0-6.3
17/1 1000-1600 140 6.4-6.5
Experiments
Technical grade alum was added as a 20% solution to the influent sewage just ahead the grit
chamber. The flow of alum was adjusted over the day so that a constant dose of alum per unit
volume of sewage was achieved. The operating conditions are presented in Table 5.
Results
Results of the full-scale experiments are presented in Table 6 and Figure 6.
If the effluent quality with the biological treatment (Table 3) is compared to the insults achieved
with the chemical treatment (Table 4, Table 6 and Figure 6), it appears that the chemical
treatment gives almost as low BOD values as the biological treatment. Furthermore, the chemical
treatment removes up to 90% of the phosphorus content of the sewage.
The results of the full-scale experiment indicate that separate flocculation basins may be omitted.
Visual observations during the experiments revealed that the flocculation pH should be controlled,
as pH over 6.7 yielded a floe with poor settling properties.
Sludge Handling
In Kiruna the normal amount of primary sludge is 50-70 g total solids per capita per day. When
alum is added to the sewage, this amount increases to about 100 g total solids per capita per day.
During the full-scale experiment the total solids concentration of the sludge fell from the normal
5-8% down to 2-3%. The sludge volume hence increased to about 4 times the normal.
The sludge dewatering equipment in Kiruna consists of a "roto-plug" gravity dewatering drum
followed by a screw press. The equipment captures 80-90% of the total solids in the primary
sludge. With chemical treatment sludges the solids capture was down to about 30%. Cationic
-------
260
PHOSPHOROUS
ppm
6-
5 -
4 •
3
2
1
ppm
BO-
100-
50
vn
minrnn OPERATION TIME
K^sSH INFLUENT
1 1 EFFLUENT
n
a
\
\
%
I
y
%
!
Y
^
y.
X
X
1
y,
y
X
/^
^
y
X
X
1
1
A
X
^
1
I2J HJ 14)
0 12 0 12 0 12
7}
y
y
Y,
^
i
y.
y
^
f**
1
faw
yW/
v$
771
\
V
X
X
^
|
n
\ ^
u
0 12 0
n \
"
/
/
/
/
/ /
^
/
/
/
/
/
/
(1
?
',
1 1 rrl
r T
12 0
y
12 0
FIGURE 6 Operation time with chemical treatment and phosphorus and BOD7 in influent and
effluent
polymers had a very good conditioning effect on the sludge, but the dynamic action of the sludge
dewatering equipment destroyed the floes formed by the polymers. It is, however, believed that
the sludge could be successfully dewatered with static-type dewatering equipment.
CONCLUDING DISCUSSION
The experiences with the activated sludge treatment have proved that biological treatment is
possible even at very low sewage temperatures. As the metabolizing activity of the activated sludge
bacteria is considerably reduced, long aeration periods, 4-5 hours, and therefore large aeration
basins, are required.
A chemical treatment is much less sensitive to low temperatures and requires only about 0.5 hour
detention time in the flocculation tanks. The difference in investment costs will in many instances
be so large that the increased running costs are justified. '
-------
261
If a community has an existing primary treatment plant with a long detention time (more than 2
hours), it may be possible to achieve a substantial increase in treatment efficiency just by a simple
addition of flocculating chemicals to the influent.
The BOD removal with chemical treatment is somewhat inferior to what can be achieved with
biological treatment. This drawback, however, is compensated by the superior phosphorus
removal.
TABLE 6
Analytical Results - Simplified Chemical Treatment
INFLUENT
EFFLUENT
Date
12/1
13/1
14/1
15/1
16/1
17/1
Time
1930-2130
1840-2140
1220-1530
1000-1400
0000-0300
0300-0600
0600-1100
0530-0730
0730-0930
0930-1130
1130-1330
Suspended
solids
ppm
120
110
130
110
60
15
52
4
7
59
98
COD
ppm
33
35
35
40
19
7
19
6
7
30
42
Time
2100-2245
2030-2230
1500-1820
1300-1500
1500-1715
0320-0520
0520-0720
0720-0920
0920-1120
1120-1400
1030-1230
1230-1430
1430-1530
Suspended
solids
ppm
24
14
40
13
17
26
20
17
33
20
8
16
16
COD
ppm
14
13
16
11
11
15
9
18
9
17
8
15
18
Aluminum
ppm
1.3
1.3
1.6
1.1
0.8
1.4
1.3
0.8
1.0
0.8
1.1
1.3
*COD was determined with potassium-permanganate oxidation
-------
262
REFERENCES
Balmer, P., Berglund, D. T. and Granstrand, G. 0964) Report to the town of Kiruna about
activated sludge experiments, I. (in Swedish).
Balmer, P., Berglund, D. T. and Widell, A. (1964) Report to the town of Kiruna about activated
sludge experiments. III. (in Swedish).
Balme'r, P., Berglund, D. T. and Enebo, L. (1967) Step-sludge - a new approach to wastewater
treatment, J. Water Pollut. Cont. Fed., 39:1021.
Gustavsson, B. and Westberg, N. (1965) Experiments with treatment of sewage from the town of
Kiruna with the activated sludge method, (in Swedish), Royal Institute of Technology
Publications. Vol. 4.
Hilme'r, A., The influence of temperature on the activated sludge process, (in Swedish), Royal
Institute of Technology Publications.
-------
BIOLOGICAL SEWAGE TREATMENT IN A COLD CLIMATE AREA
Shigeo Terashima, Keichi Koyama, and Yasumoto Magara
INTRODUCTION
Recently many sewage treatment plants using activated sludge processes have been installed in the
Hokkaido and Tohoku districts which are cold climate areas in Japan. In these plants it is
necessary to give attention to low sewage temperature in winter and thaw seasons.
The rate of run-off is often about twice dry weather flow. Thawing flow carries the turbid
substances that were piled up in winter and lowers the sewage temperature. Therefore, it is
considered that thawing flow affects sewage treatment process, but no reports about the effect of
thawing are available.
The removal characteristics of BOD and nitrogen at the Souseigawa municipal sewage treatment
plant (Sapporo, Japan) from March 1967 to February 1970, are shown in Table 1.
TABLE 1
The Effect of Temperature on Purification at the Souseigawa Plant
Sewage temp. BOD removal (%) (Org-N. + NH3-N.) removal (%)
<10°0C 82 0
11-1/C 82 23
>16 C 87 25
This shows that nitrogen removal ceases below 10° C but BOD removal is little affected by sewage
temperature. It is assumed that there is a different effect on BOD and nitrogen removal activity of
activated sludge.
The activated sludge process is composed of (1) removal of wastes, mainly by biosorption, after
contact of the waste with activated sludge, (2) removal of wastes by the anabolism and catabolism
of the absorbed substrate, (3) aeration to supply the oxygen for these reactions and, (4)
solid-liquid separation. Several reports have been published on the effect of temperature on
activated sludge processes (Banerji, 1965). Ludzack et al. (1952) showed that sewage temperature
had little effect on the activated sludge. The waste removal at 30° C was 10% higher than at 5° C,
but the solids yield was greater at 5° C than 30° C. (Keefer, 1962) reported operating results over
a 20-year period for a municipal sewage treatment plant. The variation of the average percent BOD
removal with temperature was dependent on the flow and increased as the organic load was
increased. Okun (1949) reported little effect of temperature on BOD removal in the activated
sludge process between 8-35° C.
These reports discussed the overall effects of temperature on the activated sludge process and were
not investigations of each individual process component of the total system. There have been no
investigations on compounds other than BOD.
263
-------
264
It is necessary to investigate the effect of each process and to know what process is mainly
affected by sewage temperature. It is possible to have a good activated sludge plant in a cold
climate area by strengthening that process mainly affected by sewage temperature.
We have been investigating the thawing effect on sewage and the effect of low sewage temperature
on the activated sludge process. This paper deals with the effects of thawing and the effects of low
sewage temperature on the aeration process, substrate removal, nitrification and settling
characteristics of activated sludge.
THE EFFECTS OF THAWING ON SEWAGE
Procedures
Thawing is affected by atmospheric temperature, sunshine, rainfall, underground temperature, etc.
Atmospheric temperature is considered the primary influence for run-off analyses. Degree-days
and degree-hours are used as indices of atmospheric temperature. The degree-day is defined as the
cumulative number of daily mean atmospheric temperatures above 0° C within a certain period,
and degree-hour as the cumulative number of hourly atmospheric temperatures above 0° C within
the day. Sewage drainage areas are more affected by atmospheric temperature than river
catchment areas because sewage drainage areas are very limited and thawing is directly related to
hourly change of atmospheric temperature. So, the degree-hour is used as an index of thawing
run-off analysis.
We investigated two sewage drainage districts at Sapporo, Japan. Their characteristics are shown in
Table 2.
TABLE 2
Characteristics of Drainage Districts
Souseigawa District Makomanai District
Drainage Area 623 ha (April 1968) 37 ha (April 1968)
736 ha (April 1969)
Population 125,090 (Dec. 1967)
Sewer System Combined system Separate system
Type of District Urban area Residential area
The rate and quality of sewage flow were observed at the receiving well of the sewage treatment
piant (1968) or at the final pumping station (1969) at the Souseigawa district; the same were
examined at the outlet of the storm sewer at the Makomanai district. Sewage was analyzed for
temperature, pH, 4.3 alkalinity, chloride ion (CO, total residue (T-Re), suspended solids (SS),
chemical oxygen demand (COD), biological oxygen demand (BOD), organic nitrogen (Org N),
ammonium nitrogen (NH3-N), nitrite nitrogen (N02-N), and nitrate nitrogen (NO3-N) according
to the 12th edition of Standard Methods (1965). Sampling was done at 3:30 PM - 4:00 PM
(Souseigawa district) and 2:30 PM - 3:00 PM (Makomanai district). In Sapporo, it was most
probably to find the highest atmospheric ternperature at about 2:00 PM. Therefore it was
considered that this sample was most affected by.thaw. To calculate the degree-hour, an
atmospheric temperature was summed from 3:00 AM to the next day at 3:00 AM. The rate of
thawing run-off was gained by reducing the dry weather flow (55,000 M3/d-1968. 60,000
-------
265
M3/d-l969) from the total rate of sewage flow. If there was a rainfall, the rate of storm run-off
was reduced.
Results and Discussion
The Rate of Thawing Run-off
Figures 1, 2 and 3 show the daily aspects of a degree-hour and the rate of thawing run-off. The
relationships between a degree-hour and the rate of thawing run-off are shown in Figure 4. The
next empirical formulas (1-1), (1-2), and (1-3) are attained from this relationship.
Souseigawa district
March 1 -April 6, 1968
correlation coefficient r = 0.87
Qt = 0.743AD + 17.2A
March 6 - March 26, 1969 (excluding March 21)
Qt = 0.747 A D + 27.7A
correlation coefficient r = 0.94
Makomanai district
1969 (duration of cold days)
Qt = 0.14AD + 9.7A
correlation coefficient r = 0.94
Here, Qt = the rate of thawing run-off (M3/D)
D = the degree-hour (° C hr.)
A = drainage area (ha)
I
ORATE OF THAWING RUN-OFF
• DEGREE HOUR
\
|
fe
I
(1-1)
(1-2)
(1-3)
DATE
FIGURE 1 Rate of thawing runoff and degree-hour (Souseigawa district, 1968)
-------
266
O RATE OF THAWING RUN-OFF}
• DEGREE HOUR
IS 20
DATE
FIGURE 2 Rate of thawing runoff and degree-hour (Souseigawa district, 1968)
Ik
?
I750
1500
1250
1000
o 750
in
tc
500
250
—O— RATE OF THAWING RUN-OFF (mVDAY)
--Q P.M. 3-30, RATE OF THAWING RUN-OFF (-j^ mtonW
—• DEGREE HOUR
FIGURE 3 Rate of thawing runoff and degree-hour (Makomanai district, 1969)
-------
267
•S
CO
5
P
x!2
"W
fc3
"in
a:
100
75
50
—O SOUSEIGAWA DISTRICT (1968)
—CD SOUSEIGAWA DISTRICT (1968)
CD
O
—-©— MAKOMANAI DISTRICT
o
25 50 75 100
DEGREE HOUR — "C hr
125
BO
10.0
7.5 fcS
5.0 i |
I
5
2.5
-------
200
COEFFICIENT OF CORRELATION
* 0985
v 60 CO 60 200
OBSERVED VALUE - "C hr
FIGURE 5 Correlation between observed and calculated degree-hour
TO
# 60
I
40
2
COEFFICIENT OF CORRELATION
• 036
COEFFICENT OF CORRELATION
-0*7 O
345678
WATER TEMPERATURE - "C
FIGURE 6 Correlation between water temperature and runoff percentage
-------
269
very significantly affected by thawing inflow. However, at the Makomanai district, there was poor
correlation compared to the Souseigawa district during the season. Hourly observation data
(Figure 7) shows the reason why significant correlation was not attained. The correlation at the
end of the season differed from that at the beginning. This was caused by changing sewage
temperature, influenced mainly by ground temperatures.
Suspended matter was increased up to 1,500 mg/l at the Makomanai district and up to 500-600
mg/l at the Souseigawa district by thawing. The correlation of suspended solids and percentage of
thawing run-off at the Souseigawa district is shown in Figure 8-1. The correlation of suspended
solids and run-off rate at the Makomanai district is shown in Figure 8-2. Each figure shows that
suspended solids and chemical oxygen demand attributed to suspended solids increase accordingly
with increments of thawing run-off. The results of the Makomanai district show that thawing
run-off in early periods contains much turbidity compared with final periods.
o
«
UJ
ir
UJ
te.
I
10
FIGURE 7 Hourly observation at Makomanai
-------
270
8!
I
100
80
I eo
S-
5
— 500
SUSPENDED SOLIOS-i.j/1
O Mv. 26
9.9
Mm 17
50 100 200 2SO 300
COO,, my|
020 V
O.B |:
0.10 o
u
OjQS
FIGURE 8-2 Correlation among rate of runoff and suspended solids, COD (Makomanai
district) ss
-------
271
Other components of the sewage are shown in Figure 9. The figures show that generally each
component, except chloride and nitrate, increases with temperature decrement which relates to
the rate of thawing run-off. However, the range of those fluctuations was not so large as to affect
the sewage treatment facilities.
9
8
7
6
246
WATER TEMR--C
2 4
WATER TEMP.-*C
20
10
246
WATER TEMP.-»C
_ 1.0
f
I
^> 0.5
£46
WATER TEMR-°C
-0.08
E
I
fo.04
f 2
o
246
WATER TEMR-"C
246
WATER TEMP.-'C
20
10
40
3
246
WATER TEMR-°C
2 4 6
WATER TEMR — "C
FIGURE 9 Various sewage components
-------
272
Conclusion
The rate of thawing run-off significantly relates to the degree-hour. If these correlations are known
for a certain drainage area, it is possible to predict the rate of thawing run-off from the highest and
lowest atmospheric temperatures of the day. However, these correlations become less obvious at
the end of the thaw season.
Sewage temperature and suspended solids are significantly affected by thawing run-off, especially
when sewage temperature drops below 5° C.
THE EFFECT OF LOW SEWAGE TEMPERATURE
ON THE ACTIVATED SLUDGE PROCESS
The Effect of Sewage Temperature On Oxygen Transfer
Procedures
The effects of temperature on oxygen transfer in a diffused aeration tank were investigated by
measuring the oxygen transfer coefficient at every 5° C decrement from 25° C to 5° C under a
fixed rate of air flow.
Completely mixed aeration tanks were used, with porous plate diffusers installed in the bottom.
The aeration tank is shown in Figure 10. The oxygen transfer coefficient was obtained by a
non-steady state procedure {Eckenfelder, O'Connor, 1961) under stabilized operating conditions.
The aeration tank was filled to a desired volume with tap water. Dissolved oxygen in the water was
removed by admitting nitrogen gas. After the dissolved oxygen approached 1 mg/l, air was
admitted at 10 l/min or 20 l/min, and the dissolved oxygen was continuously measured by an
oxygen analyzer (Toshiba-Beckman Co. Ltd., 777 type).
Oxygen transfer can be expressed by Formula 2, using an overall oxygen transfer coefficient.
dC
— = KLg (Cs-C) (2)
Here, Kj_a = overall oxygen transfer coefficient
C = dissolved oxygen concentration
Cs = saturated dissolved oxygen concentration
The transfer coefficients computed from the above relationship are shown in Table 3 and Figure
11. These results show that the oxygen transfer coefficient is affected by temperature. This effect
can be expressed by Formula 3:
-------
273
in which
T = temperature (° C)
K|_a T = transfer coefficient at temperature T
KLa jo = transfer coefficient at 20° C
0 = temperature coefficient
The temperature coefficient, which is computed from Formula (3), was found to be 1 017 This
value agreed favorably with the result by Eckenfelder (1961).
SUBSTRATE
STORAGE
COOLING
APARATUS
AERATION SEDIMENTATION
TANK TANK
(Complete \ Volum«-63L
SLUDGE
RETURNING
PUMP
LARGE SCALE WATER BATH
FIGURE 10 Complete mixing continuous flow units
TABLE 3
The Effect of Temperature on Oxygen Transfer Coefficient
Oxygen Transfer Coefficient (I/hour)
10 1/min
Water
Temperature
10° C
15° C
20° C
25 C
2.6
3.0
3.2
3.5
3.2
20 1/min
5.2
6.5
6.1
6.9
7.4
The ratio of the mass of oxygen transfer at any water temperature to the transfer at 20° C is
calculated from Formulas 2 and 3, assuming the dissolved oxygen concentration is to be held
constant, for instance, at 2 mg/l. This value, shows, as in Figure 12, that the mass of oxygen
transferred increases with water temperature decrement. The mass of oxygen transfer is more
-------
274
affected by increasing a saturated dissolved oxygen concentration than by decreasing an overall
oxygen transfer coefficient. From this result, there need be no concern that oxygen transfer will
be hampered by a lowering of the water temperature.
LJ
O
O
O
a.
LU
FLOW RATE
ZOL/min
FLOW RATE
lOL/min.
0 5 10 15 20 25
WATER TEMPERATURE-*C
FIGURE 11 Oxygen transfer coefficient and temperature
o
°
e
tc.
u
u.
CO
LJ
CO
CO
o
I
1.2
1.0
0.9
0 5 10 15 20 25
WATER TEMPERATURE - °C
FIGURE 12 Ratio of mass of oxygen transfer
-------
275
The Effects of Temperature On the Purification Activity of Activated Sludge
Procedure
Experiments shown in Table 4 were performed to find the effect of temperature on the activated
sludge process.
Run No.
1-1
1-2
1-3
2-1
2-2
2-3
2-4
3-1
3-2
3-3
3-4
4-1
4-2
5-1
5-2
5-3
Substrate
sewage
91
skim milk
skim milk
+ peptone
skim milk
glucose
TABLE 4
Experimental Details
Aeration tank Aeration period
M.C.A.T.1 6 hr
" »
" »»
C.M.A.T.2
M.C.A.T.3
Water temp.
20: c
15° C
10° C
20° C
15" C
10° C
20° c
15° C
10° C
5°C
20° C
10° C
20" C
15° C
10° C
Operation
C.F.,4 1440 1/D
C.F., 10801/D
C.F., 1290 1/D
C.F., 7041/D
C.F., 1440 1/D
Batch
multi-cell (6 cells) aeration tank
completely mixed aeration tank
(3) multi-cell (14 cells) aeration tank
(4) continuous flow
Apparatus. The apparatus used in the experiments is shown in Figure 13. A plug flow aeration
tank was used in runs No. 1 and No. 3. This aeration tank was simulated from an actual aeration
tank of a sewage treatment plant by dividing it into multi-cells.
Substrate. The model plant (run No. 1), which was fed domestic sewage, was installed in the
Makomanai sewage treatment plant. Preclarified sewage was used as a substrate, and the model
plant fed under load fluctuation, which caused only fluctuation of sewage concentration.
Synthetic sewage (runs No. 2 and No. 3) had ammonium chloride added to supply nitrogen to the
activated sludge.
Operation. Model plants were normally operated at 1 l/min. inflow and 0.25 l/min. return sludge.
Under these loadings, aeration periods were 6 hours and sedimentation periods 1 hour. However,
when suspended solids were carried over from the sedimentation tank, and the aeration tank could
not sustain the desired activated sludge concentration, the rate of inflow was decreased. The
volume of return sludge was then increased so that the aeration period would remain constant at 6
hours.
-------
276
Acclimation was performed by increasing an organic load from low values to a set value. Domestic
sewage sludge was used for seed. Later, to confirm acclimation, testing was started. Batch
experiments were performed using continuous flow plant sludge.
Analysis. The sample was analyzed for water temperature, pH, dissolved oxygen, total residue,
suspended solids, volatile suspended solids, chemical oxygen demand, biological oxygen demand,
organic nitrogen, ammonium nitrogen, nitrite nitrogen; nitrate nitrogen and sludge volume index
(SVI) according to the 12th edition of Standard Methods (1965).
Results and Discussion
The results of continuous flow plants are shown in Table 5 and the results of batch experiments
are shown in Table 6.
Settling Characteristics. The sludge volume index (SVI) proved that sludge settling characteristics
depend strongly on the sewage temperature, as in Figure 14.
TABLE 5
Results of Continuous Flow Plant
1-1 1-2 1-3 2-1
Water temp. C
Organic load
MLSS2
SVI
COD2
Org-N+NH3-N2
COD2 ,
CODs3'2
20
0.25
3778
90
286
55.9
15
0.35
2995
178
6 20
0.38 0.34
2836 3035
181 45
346 365 322
63.8 67.3 34.4
Run No.
2-2 2-3
20 15
0.56 0.17
866 2336
- 322
Influent
152 138
10.0 10.4
Effluent
2-4 3-1
10
0.15
2431
402
233
17.4
20
0.33
2636
51
275
27.5
3-2
15
0.47
1000
112
157
15.7
3-3 3-4
10 5
0.25 0.14
2647 3219
345 250
161
16.1
Org-N2,
NH3-N2
NOj-N2
NO3-N2
COjD removal
COD removal5
Org-N+NH3-N
removal %
NO2-N-t-NO3-N
removal %
Nitrification
Activity6
1) Organic load = kg COD/kg MLSS/day
2) Concentration = mg/1
3) CODs = Soluble COD
14
135 145
58
17
17
19
6-g 2.JJ l.g 0.7 1.1 0.
xlO"4 xlO xlO xlO'^ xlO"
147
14.7
83
45
43
2.1
18.6
7.2
1.4
71
89
327
63
45
31
22
1.8
34.6
3.3
1.2
87
91
414
43
274
37
209
17.6
35.5
0
0
25
90
128
22
51
31
23
-
- .
-
85
91
394
-
7.1
1.1
2.1
11.4
.
-
.
18
9
8
26
1.2
0.2
0.2
14.9
93
94
167
87
115
13
117
15.8
7.8
0
7.1
51
94
83
0
52
17
51
2.6
8.9
0.36
4.4
81
96
292
58
27
11
11
2.6
7.5
2.1
0.7
83
93
188
29
26
9
15
4.6
2.3
0
3.2
83
94
192
57
45
7
32
1.2
3.8
0
0
69
96
144
53
0
0
'* xlO'2 xlO"12
'4) Calculated from CODs
,5) COD removed = g COD/day
^6) Nitrification activity = (N02 -N+NO3 -N) MLSS day
-------
-600-
o
»
AIR
M50-»f- 3OO-
277
INFLOW [
RETURN SLUDGE
T+
•0
LU
z
1
10
^
10
r-
_i
*•
_i
£>
¥ U
_i
r-
1
-------
278
HEATER
DIFFUSER
FLOW METER
FIGURE 13-3 Batch unit
Here, e = void ratio
Re = Reynold's number
A = coefficient of flow properties
Figure 15 shows the change of SVI and substrate concentration after substrate contact with
activated sludge. SVI increased in gradual proportion to absorbed substrate and showed maximum
values at the time when substrate was almost removed from the solution. Afterward, SVI gradually
decreased with the utilization of absorbed substrate. Therefore, it was considered that SVI was
related to the absorbed substrate.
It is considered from the above description that the high value of SVI at low water temperature is
not only attributable to change of water viscosity but also to changes of activated sludge floe
physical properties like density, bound water ratio, and surface electric charge. Further
investigations into these relationships between the biological and physical properties are necessary.
COD. The removal ratio of soluble COD was not affected, but COD that contained suspended
matter was affected, by water temperature in a continuous flow aeration unit, as in Figure 16. This
showed an increase of suspended matter, carried over from the sedimentation tank in low water
temperatures. Total mass of COD removed decreased with water temperature, as in Figure 17. This
was the result of floe carry-over at the Makomanai model plant; other model plants were operated
under low organic load to repress the high SVI or low overflow rate of the sedimentation tank to
prevent floe carry-over.
-------
279
Water temp.
Organic load*
MLSS mg/1
COD initial (mg/1)**
COD 6 hrs (mg/1)
TABLE 6
Results of Batch Experiments
K
4-1
20° C
0.34
3500
278
54
2.5
xlO
xibOJ*
0.18
3500
139
49
2.2
xlO
6.5g
0.40
3000
286
44
0-9
xlO"2
4.08
xlO'*
Run No.
4-2
10° C
0.30
2000
66
18
l^
3.72
xlO"4
5-1
20° C
0.18
2200
8.7
xlO4
3.9
«ir» *
5-2
15° C
0.18
2200
xlO:?
2 3
5-3
10° C
0.18
2200
xlO^
14
•Nit.
* Organic load = kg COD/kg MLSS/Day
** COD initial was determined after 3 min. of aeration K = I/day
The removal of COD is shown in Formula 5:
dL
dt
Here, L = substrate concentration
S = initial activated sludge concentration
K = substrate removal rate
= - KSL
(5)
400
350
300
250
- 200
CO
150
100
50
0
O RUN No I
X RUN No. 2
RUN No. 3
5 10 15 20
WATER TEMPERATURE-"C
FIGURE 14 SVI and water temperature
-------
280
Soluble COD removal rate in batch tests (Table 6) showed it decreased with sewage temperature.
Temperature coefficient (6) which was obtained according to Formula 6 is 1.089.
(6)
Here. KT = the rate at T° C
°
= the rate at 20° C
T = water temperature
6 = temperature coefficient
This (0) showed that the removal rate was about half when the water temperature was decreased
by 10° C. Figure 16 shows that the removal rate was not affected by temperature in a continuous
flow aeration tank as in the batch experiments. This was due to the effluent of continuous flow,
sampled after 6 hours of aeration.
There is a relationship between settling characteristics and absorbed substrate of activated sludge
that is shown in Figure 15. And the removal rate is affected by absorbed substrate of return
- 300
ef
o
u
- 200
>
n
12
AERATION PERIOO-hr
CONDITION OF EXPERIMENT
Wofcr temperoiure « 20»C Substrate • skim milk, peptone
A- Organic load « 0.50 kq
-------
281
sludge. From these points, it is assumed that the stabilization time for oxidation or synthesis of
absorbed substrate (that is, the aeration time after the removal by absorption) is the key to
successful operation of the activated sludge process.
Nitrogen. The primary nitrogen components of influence to sewage treatment plants are organic
nitrogen and ammonium nitrogen. These components are used by bacteria and converted to cell
nitrogen or to nitrite and nitrate nitrogen. Nitrification is very sensitive to dissolved oxygen in
mixed liquor, so results in Tables 5 and 6 are from experiments performed above 3 mg/l of
dissolved oxygen.
The results of batch units (Fig. 18} show that the organic and ammonium nitrogen removal are
first order reactions, expressed by Formula 5, and nitrification is a zero order reaction, shown by
Formula 7.
dt
= KS
(7)
Here, L = nitrite and nitrate nitrogen
S = mixed liquor suspended solids
K = rate of nitrification
The removal rates of organic and ammonium nitrogen and the rate of nitrification (Table 6),
determined according to Formulas 5 and 7, show that these rates decrease with temperature. The
temperature coefficient (9) according to Formula 6 is 1.092 for organic and ammonium nitrogen
and 1.106 for nitrification.
The experimental results of continuous flow plants in Figure 19 show the same phenomena as in
municipal sewage treatment plants of the same organic load. When the temperature is decreased to
5° C, nitrification does not occur and the nitrogen compounds in the influent are discharged
without stabilizing.
T
o
i
_j
5
2
UJ
Lt
0
s
100
90
80
TO
GO
50
40
30
20
10
0
COD
v*-*^.
*x>/!/"xb
// O RUN Na '
'
/ X RUN No. 2
/ • RUN No 3
d
— •— TOTAL COO
— O~ SOLUBLE COO
5 10 IS 20
WATER TEMPERATURE-«C
FIGURE 16 COD removal ratio and water temperature
-------
282
From the above, the rate coefficient of nitrogen removal or nitrification is smaller and more
sensitive to temperature than that of COD, The nitrogen removal reaction is not completed in 6
hours of aeration as is COD, and the nitrogen compounds in the influent are discharged with
partial or no stabilization from the aeration plant.
a
o
o
400
300
200
100
0 5 10 15 20
WATER TEMPERATURE-*C
FIGURE 17 Mass of COD removed and water temperature
WATER TEMP
20* C
WATER TEMP
0 48 12
AERATION PERIODS- HOUR
FIGURE 18-1 Effect of water temperature on nitrogen decrement
-------
283
IT19/I
0 4 8 12
AERATION PERIODS-HOUR
FIGURE 18-2 Effect of water temperature on nitrogen removal
I
UJ
O RUN No. I
X RUN Na2
RUN No. 3
5 10 15 20
WATER TEMPERATURE -»C
FIGURE 19 Nitrogen removal rate and water temperature
-------
284
CONCLUSIONS
The following conclusions have been reached from these studies:
1. At low temperatures, the oxygen transfer coefficient in the aeration tank decreases, but the
mass of oxygen transfer increases slightly with the lowering of temperature.
2. Sludge volume index increases at low temperatures. It creates a high COD effluent.
3. The value of SVI changes with time after substrate contact with activated sludge. At a low
sewage temperature, it is necessary to take sufficiently prolonged aeration time to stabilize
the activated sludge which absorbed substrate.
4. The removal ratio of soluble COD is not affected by a decrease of sewage temperature, but
total COD of effluent increases due to an increase of suspended matter.
5. The rate of removal of soluble COD varies with water temperature. This relationship is shown
in Formula 6, and the temperature coefficient is 1.089.
6. Nitrogen removal is influenced remarkably by temperature. The rate of removal of organic
and ammonium nitrogen and the rate of nitrification are shown by Formula 5 and Formula 7
respectively. These rates decrease with lower temperatures, and temperature coefficients are
larger than that of removal of COD. This is the reason the effluent of low sewage temperature
contains high nitrogen.
ACKNOWLEDGMENT
A part of this research was supported by a scientific research grant of 1969 from the Department
of Education of Japan. The City of Sapporo supported this study with convenience and
expenditure.
REFERENCES
Banerji, S. K. (1965) Biological removal of colloidal waste in activated sludge process, Ph.D. thesis
of U. of Illinois.
Eckenfelder, W. W. and O'Connor, D. J. (1961) Biological Waste Treatment, Pergamon Press.
Keefer, C. E. (1962) Temperature and efficiency of the activated sludge process J W P C F Vol
34. '
Ludzack, F. J., Schaffer, R. B. and Ettinger, M. B. (1952) Temperature and feed as variables in
activated sludge performance, S.I.W., Vol. 24.
Okun, D. A. (1949) System of bioprecipitation of organic matter from sewage, S.W.J., Vol. 21.
Sakai, T. (1963) A study of the snow-melt runoff of rivers (Japanese), Transactions J.S.C.E No
95.
-------
285
Standard Methods (1965) 12th edition.
Tanbo, N. and Abe, S. (1969) Behaviors of floe blankets in a upflow clarifier (Japanese),
J.J.W.W.A., No. 415.
-------
MICROBIOLOGIC INDICATORS OF THE EFFICIENCY
OF AN AERATED, CONTINUOUS-DISCHARGE,
SEWAGE LAGOON IN NORTHERN CLIMATES
John W. Vennes and Otmar 0. Olson
INTRODUCTION
Sewage stabilization lagoons have been used in the United States as a method of waste treatment
for the past 20 years. The first designed sewage stabilization lagoon in North Dakota was approved
by the State Department of Health in 1948 (Van Heuvelen, et al., 1960). The design criteria
established for this system were much the same as present standards of 20 Ibs. BOD5/acre/day
with a minimum of 100 days retention for winter storage.
The BOD5 loadings that can be adequately stabilized under aerobic conditions, with oxygen by
algal photosynthesis being the primary contributor to aerobiasis, depend on solar radiation,
temperature, sewage characteristics, wind action and depth. Temperature and solar radiation are of
prime consideration as regards the growth of algae and bacteria. Mixing and distribution of waste
in a conventional lagoon are primarily dependent upon wind action. A common problem in
conventional lagoons occurs during the first warm days in the spring when anaerobic conditions
prevail due to benthic deposits rising to the upper layers of liquid. The recognition of problems
encountered in maintaining aerobic conditions in these conventional lagoons led to the application
of artificial aeration.
The aerated lagoon is one of the most recent developments in the biological waste treatment
system. This concept was initially developed to supplement oxygen during the period of spring
break-up by artificially aerating stabilization ponds. Lagoon aeration technology has been
developed to the point where aerated lagoons are now designed for complete waste treatment
(Sawyer, 1968).
METHODS
The city of Harvey, North Dakota, treats its domestic waste in an aerated lagoon system. The
system consists of two cells, each having a water surface area of 1% acres. The cells are designed to
operate in series, each providing a 20-day retention time. The dikes have a 3 to 1 slope and support
a 10-foot berm, which allows the complete system to be constructed in less than 6 acres. Overflow
manholes in the two cells allow them to be operated at a 10-foot depth with a
continuous-discharge to the Sheyenne River.
The aeration system consists of two 15 hp Sutterbilt blowers, each capable of providing 270 cfm
at 9 psi. The air is distributed to the two cells through a plastic header pipe paralleling the cells
with weighted diffuser tubes connected to the header. There are 78 diffuser lines distributing air
into the primary cell and 40 lines in the second cell, giving a total of over 15,000 feet of tubing
disbursing the air. The equipment was supplied by Hinde Engineering Company, Highland Park,
Illinois, and is known as an Air-Aqua System.
286
-------
287
The blowers are operated alternately with a time clock to assure continuous operation of the waste
treatment facility. The blowers have been in essentially continuous operation since October 1965.
Ice cover usually is seen in the secondary cell in November or early December while the primary
cell does not show ice cover until late December or early January.
The primary cell was designed to be loaded at approximately 400 Ibs. of 5-day biochemical oxygen
demand (BODs)/acre/day and the secondary cell at about 100 Ibs. of BODS/acre/day. Each cell
has a detention time of about 20 days.
Sampling Procedure
Sampling of 24-hour composite raw city waste and effluent from the primary and secondary cell
manholes was carried out during the period from January 1966 through June 1969. Daily flows of
sewage averaged 175,000 gpd with changes occurring during spring run-off and rainy weather.
Samples were usually collected on a monthly basis, however, on occasion two samples were taken
during the same month. Since this facility is about 150 miles from the laboratory, the samples
were maintained at approximately lagoon temperature in an insulated container and transmitted to
the laboratory the same day. Microbiologic determinations were carried out immediately on arrival
at the laboratory while the determinations for other parameters were made either the same day or
the following morning. Samples in the laboratory were maintained at 5° C prior to the following
day determinations.
Laboratory Determinations
Determinations for coliform, fecal coliform, and enterococci were made by the Millipore filter
technique while the total bacteria populations were enumerated with tryptone-glucose extract agar
(Standard Methods, 1965). The determinations for BOD5, total and suspended solids, pH and
Kjeldahl nitrogen were made according to Standard Methods (1965). A few determinations were
made for biotin by the bacteriologic assay method (Wright and Skeggs, 1944) with Lactobacillus
plantarum. Several determinations for sulfide and phosphate were also made by using Standard
Methods (1965).
RESULTS
Laboratory determinations for BOD5, Kjeldahl nitrogen, pH, total and suspended solids, coliform,
fecal coliform, enterococci, and total bacteria are tabulated in the Appendix (Tables 5-12).
Graphic presentation of this data is seen in Figures 1 through 6.
This aerated lagoon discharged its effluent continuously into a water course; therefore it was
considered useful to determine the mean values for several parameters monitored during the period
of the study. It can be seen in Table 1 that the mean BOD5 for the period was 53 mg/liter (S.D.
±45 mg/liter). The suspended solids show a mean value of 67 mg/liter (S.D. ±39 mg/liter). Mean
nitrogen shows values of 7.7 mg/liter (S.D. ±3 mg/liter). There was considerable variation in the
composite raw samples in that the mean BODS of the 24-hour composite raw was 332 mg/liter
(S.D. ±422 mg/liter). It is thus apparent that the loading on the primary cell changed rather
drastically during the period of the study due primarily to a milk plant.
-------
288
TABLE 1
Statistical Correlation of Several Lagoon Parameters
mg/liter
Parameter Raw Primary Secondary
BODS Mean 332(263)* 126 53
S.D. 422(134) 76 45
TVS Mean 567(496) 401 353
S.D. 424(118) 94 98
VSS Mean 191(166) 118 67
S.D. 168(118) 74 39
N2 Mean 20(15.5) 11.2 7.7
S.D. 27(5.3) 3.8 3.0
"Value obtained by disregarding one apparently non-representative sample.
An attempt was made to assess the effect of changing loadings and temperature on this lagoon by
comparing BODS values with coliform, fecal coliform, enterococci, and total bacteria. Figures 7
through 9 display this relationship on a log-log basis and suggest that microbial populations in the
secondary lagoon are directly related to the BODS or total loading.
Correlations of the lagoon parameters tabulated during the different periods of the study, namely,
for the complete sampling period, for the months when the lagoon liquid temperature was near 0°
C (i.e. January through March), and the period when the lagoon was near 10° - 20° C (i.e. June
through September) are displayed in Table 2. It can be noted in this table that there is a
reasonably good correlation between the BODS and the Kjeldahl nitrogen during all periods of the
test. One may also observe that the BODS of the secondary lagoon, during times when the lagoon
temperature was near 0° C, showed excellent correlation with coliform and fecal coliform bacteria.
Less correlation was seen with the enterococci and BOD5, and little correlation was noted between
BOD5 and total bacteria. During the summer months, when the lagoon temperatures were between
10° and 20° C, the relations between BODS and coliform, fecal coliform, and enterococci were
less attractive. There was, however, during this time period, a reasonably good correlation between
the secondary lagoon BODS and the total bacterial population.
The relations between nitrogen and coliform, fecal coliform, and enterococci during the winter
months of January through March showed excellent correlation; however, the total bacterial
population does not show a direct relation to the total nitrogen in the secondary lagoon. Again, as
with the BODS and microbial numbers at 10° - 20° C, there is little correlation between the total
nitrogen and enteric bacteria at summer temperatures. There is some correlation between total
nitrogen and total bacteria during the summer months.
One of the apparent difficulties with the aeration system, as utilized in this continuous-discharge
lagoon, is the fact that precipitation of insoluble salts tends to decrease the porosity in the diffuser
tubes and pressures tend to build jn the system. Occasional cleaning of the system with
hydrochloric acid to solubilize the salts usually relieves the pressure and increases the pore size. It
was noted during the treatment year from January 1966 through January 1967 that excellent
reductions in BODS were obtained in the lagoon. However, during the treatment year 1967
-------
289
TABLE 2
Correlation Coefficient of Several Lagoon Parameters
Parameters Compared Correlation Coefficient
Months compared
A11 Jan. - Mar. June - Sept.
BOD-nitrogen 0.89 0.89 0.82
BOD - coliform 0.33 0.93 0.10
BOD - fecal coliform 0.13 0.93 0.31
BOD - enterococci 0.73 0.70 0.19
BOD - total bacteria 0.52 0.35 0.79
Nitrogen - coliform 0.48 0.91 0.20
Nitrogen - fecal coliform 0.12 0.96 033
Nitrogen - enterococci 0.41 0.90 053
Nitrogen - total bacteria 0.43 0.17 0 29
through 1968 a cyclic effect in the secondary lagoon parameters was noted and decreased
efficiency (due to apparent lack of delivered oxygen) resulted in rather large amounts of BODS
and microbial numbers being discharged to the water course. For example, during the months of
January through March in 1968 there was approximately a two-fold increase in the nitrogen and
suspended solids in the secondary lagoon. There was nearly a four-fold increase in the BODS.
There was also an increase of from 1 to 2 logs in the coliform, fecal coliform, and enterococci in
the secondary effluent. No consistant increase in the total bacterial population was detected.
DISCUSSION
Two of the most important physical factors in this continuous-discharge lagoon affecting the
stabilization of organic molecules, and hence microbial growth, appear to be temperature and
delivered oxygen. Since two of the other parameters related to microbial growth are substrate
availability and pH, and since it is assumed that raw sewage will usually support adequate growth
of microorganisms for near complete stabilization of the organic species present, and since pH
varied little from neutrality, the major concern with lagoon efficiency must remain with
temperature and oxygen.
This study allows the determination of these effects of temperature and oxygen on biologic
stabilization and reflects these findings in the relative abundance of several microbial species and
BODS. For example, it was noted that at near 0° C the total microbial population did not
correlate well with BOD5 reductions, while reductions in coliform, fecal coliform, and enterococci
were directly related to the BODS of the supporting medium. Initial populations of enteric
organisms were thus determined by the strength of the raw sewage. This relationship does not
remain during summer temperatures of 10° - 20° C. It would thus appear that the rates of
utilization of lagoon substrates by enteric organisms are different at different temperatures.
However, since the total microbial population varied little at different BODS loadings it may be
-------
290
that changes or die-off rates of coliforms may represent primarily the variable of time. Alternately
it may be that BOD5 is not a useful parameter in assessing relations of enteric organisms to organic
loading.
In order to verify the assumption that temperature, pH, and substrate availability did not prevent
microbial oxidation it appeared useful to compare those periods when the lagoon was known to be
at or very near 0° C with those times when it was at 10° - 20° C (when the reduction in organic
species, as reflected by the BOD5, is expected to be at its most active level).
The mean BODS of the raw composite waste during the months of January through March in the
years of 1966, 1967, and 1969 was 312 mg/liter (S.D. ±180 mg/liter), while the effluent from the
secondary cell had a mean BODS of 50 mg/liter (S.D. ±23 mg/liter). Assuming flows to average
175,000 gpd, then, as seen in Table 3, about 445 Ibs BOD5/day were loaded in the primary cell
and about 188 Ibs BODs/day were loaded in the secondary cell. A reduction of 84% in BODS was
thus noted for a period when temperatures were near 0° C.
The mean BOD5 of the raw composite waste during the months of January through March in 1968
was 665 mg/liter (S.D. ±950 mg/liter), while the effluent from the secondary cell had a mean of
160 mg/liter (S.D. ±19 mg/liter). Thus the loadings on the primary cell were about 970 Ibs
BODs/day, while 329 Ibs BODs/day were loaded on the secondary cell. An overall reduction of
75% was obtained in the total BODS.
TABLE 3
Loadings in Primary and Secondary Aerated Lagoons at Various Times
Loadings (Pounds BODj/day)
Year/s
(months)
1966-67-69
(Jan - Mar)
1968
(Jan - Mar)
All
(Jan - Sept)
All
(All)
1966-67-69
(Jan - Mar)
1968
(Jan - Mar)
All
(Jun - Sept)
Primary
455
970
408
485
Raw
312 (±180)
665 (±950)
280 (±117)
Secondary
188
329
136
265
BODS (mg/1)
Primary
129(±10)
226(±65)
93(±28)
Effluent
73
233
41
78
Secondary
50 (±23)
160 (±19)
28 (±23)
-------
291
However, during January through March in 1968, an average 737 Ibs BOD5/day were removed by
the aerated cells, while during the other years studied a mean of 382 Ibs BODs/day were removed.
Of course some of the removal of BODS in the primary cell was probably due to sedimentation,
and comparing the secondary cell removal rates may be more relevant to this study. In 1968 the
secondary cell removed an average of 115 Ibs BODs/day, while during the other years of the study
a mean of 96 Ibs of BODs/day were removed during the January through March time period, tt
appeared at this point that temperature did not determine the rate of removal of BODS, since
about equal amounts were removed during all time sequences in the secondary cell at or near 0° C.
The mean BOD5 of the raw composite waste during the months of June through September for all
years studied was 280 mg/liter (S.D. ±117 mg/liter), while the effluent from the primary cell had a
mean of 93 mg/liter BOD5 (S.D. ±28 mg/liter) and the secondary cell effluent showed a mean
BOD5 of 28 mg/liter (S.D. ±23 mg/liter). Therefore during this time period the primary cell was
loaded at 408 Ibs BOD5/day and the secondary cell at 136 Ibs BODs/day with 41 Ibs BOD5/day
being discharged to the water course. Overall reductions in BODS were about 90%, with 67%
occurring in the primary cell. It is interesting to note that the secondary cell of this aerated lagoon
removes between 95 and 115 Ibs BOD5 at all temperatures. Thus, as suggested above, temperature
was not the limiting parameter in total BOD5 reduction.
The aeration system used in this continuous-discharge lagoon utilizing a 15 hp blower at 270 cfm
at 9 psi delivers oxygen sufficient to satisfy approximately 375 Ibs BOD/day (367 to 382 Ibs
BOD/day range) at all temperatures. Thus the system is capable of removing approximately 1 Ib of
BODs/HP hr. Loss of efficiency in the system by decreased porosity in the plastic tubing is
probably not as important as the lack of diffusibility inherent in the design of the system.
Additional oxygenation transfer will be required to determine the ultimate efficiency of this
system at higher temperatures. It may be that at 0° C the maximum efficiency has already been
established by the data presented in this report. If this proves to be a fact, then a basis for aeration
capabilities of these lagoons has been further validated.
Allowing the assumption that the maximum efficiency for this system has been established, then
coliform, fecal coliform, and enterococci reflect reasonably well the biological degradation of
organics at 0° C as contrasted to summer temperatures (10° - 20° C) where little or no correlation
exists. Additional parameters will need to be investigated in order to determine those reactions
that better reflect the biologic reactions that occur at all temperatures. Since only 1% or less of the
total microbial population present in the lagoon is represented by the enteric organisms studied
here, it is apparent that other organisms must be studied to better define the role of microbiologic
indicators in the efficiency of this sewage treatment system.
One system is being studied in our laboratory which relates to several organisms that thrive in
sewage oxidation lagoons. It is concerned with the production and utilization of the B vitamin,
biotin.
We have found (Fillipi and Vennes, 1970) that biotin is produced by a number of enteric
organisms, Aerobacter aerogenes being one of the most productive. Utilization of the vitamin
seems to be limited to photosynthetic organisms (although some bacteria have been reported to
utilize or at least degrade biotin) and in our study Chlorella and probably other algae and several
species of purple sulfur bacteria seem to be the most active.
-------
292
Although production of the vitamin has not been established to be directly related to organic
loading of the lagoons, utilization seems to be associated with the production or growth of algae.
Data are not sufficient at this time to present a definitive outline of these relations, although Table
4 shows the tabulations in lagoons in eastern North Dakota. It is apparent that no good correlation
exists between BOD5 and biotin in these lagoons. Rather it is suggested, based on our previous
findings (Fillip! and Vennes, 1970), that the precursors for biotin production are present in
lagoons in different amounts and growth of biotin-producing organisms predominate on this and,
we expect, other bases. Utilization however seems to be limited primarily to the presence of algae
and thus disappearance of biotin may prove to be directly related to algal numbers.
TABLE 4
Date
9 Apr 69
5 Jun 69
21 Apr 69
21 May 70
5 Jun 70
5 Jun 69
10 Jul 69
26 May 70
20 May 70
3 Jun 70
20 May 70
3 Jun 70
20 May 70
3 Jun 70
14 May 69
11 Jun 69
25 Jun 69
10 Jul 69
7 Aug 69
Biotin and BOD5 Relations in Several Oxidation Lagoons
Source
Harvey I
Harvey Secondary
Harvey L __„
Harvey Secondary
Grand Forks Secondary - £
Grand Forks Secondary - W
Grand Forks Secondary - E
Grand Forks Secondary - W
Grand Forks Secondary - E
Grand Forks Secondary - W
Lakota Secondary
West Fargo Secondary
East Grand Forks Secondary
Crookston Secondary - W
Crookston Secondary - W
Crookston Secondary - E
Hillsboro Secondary
Hillsboro Secondary
Northwood Secondary
Northwood Secondary
Grafton Secondary
Grafton Secondary
Grafton Secondary
Grafton Secondary
Grafton Secondary
BODS
(mg/1)
27
33
58
56
119
82
166
82
91
39
12
43
35
310
166
331
20
37
27
30
513
460
180
169
156
Biotin
(ng/1)
10
20
22
134
326
120
1720
2070
2200
1950
18
53
24
2275
2950
2840
72
284
22
83
936
3530
8500
670
32
-------
293
It is suggested then that although BOD5 and total bacterial numbers are somewhat indicative of
utilization of organic substrates in lagoons, more sophisticated studies are needed to interrelate
microbial numbers and organic constituents responsible for stabilization of wastes. Biotin
production and utilization may be an example of the latter relationship.
SUMMARY
1. The limiting parameter in BOD5 reduction in an aerated, continuous-discharge lagoon in North
Dakota appears to be delivered or utilizable oxygen. No difference in BOD5 reduction was noted
in the secondary lagoon at temperatures varying from 0° to 20° C.
2, Coliform, fecal coliform and enterococcal numbers in the secondary lagoon during winter
temperatures of near 0° C were directly related to BODS and total nitrogen. During summer
temperatures little correlation between these enteric organisms and BODS and total nitrogen was
noted. However, there was a correlation between the total microbial population and BODS at
summer temperatures.
3. Although the physical aspects of the aeration system can apparently be improved by additional
engineering, the understanding of the biologic system will require extensive study of specific
ecologic relationships in the lagoon (biotin production and utilization were used to illustrate the
complexity of this system).
ACKNOWLEDGMENTS
The technical assistance of Janice Granum and Gordon Fillip! is gratefully acknowledged. The
continuing interest and assistance from the North Dakota State Department of Health and
particularly Raymond Rolshoven is also acknowledged.
This investigation was supported in part by the North Dakota Water Resources Institute with
funds provided by the U. S. Department of Interior, Office of Water Resources Research under P.
L. 88-379.
REFERENCES
Fillipi, G. M. and Vennes, J. W. (1970) Biotin production and utilization in certain natural
environments, Bact. Proc., p. 18.
Sawyer, C. N. (1968) New concepts in aerated lagoon design and operation. Adv. in Water Quality
Improvements, 1:325-335.
Standard Methods for the Examination of Water and Wastewater, 12th ed., Orland, H. P. (ed.)
Amer. Pub. Health Assoc., Inc., New York.
Van Heuvelen. W., Smith, J. K. and Hopkins, G. J. (1960) Waste stabilization lagoons - design,
construction and operation practices among Missouri basin states, J.W.P.C.F., 32:909-917.
Wright. L. D. and Skeggs, H. R. (1944) Determination of biotin with Lactobacillus arabinosus.
Proc. Soc. Exp. Biol. Med., 56:95-98.
-------
1000
100*
10
o o o
1000
• 8
o • •
100
A
a
10
JAN APRIL JULY OCT IJAN APRIL JULY OCT JJAN APRIL JULY OCT IJAN APRIL
1966 ; * 1967 * 1968 * i969
FIGURE 1 BODS changes in raw-composite, primary and secondary aerated lagoons
-------
io2
g io1
0°
I-
z
0°
APRIL JULY OCT
1966
IJAN
4—
APRIL JULY
i»«r —
OCT
-I
JAN APRIL My
1968 —
OCT
1
|JAN APRIL JULY
1969
FIGURE 2 Total nitrogen changes in raw-composite, primary and secondary aerated lagoons
N>
CO
Ol
-------
!0'
^^
1
|
g |Cf
S
1
s
i
ID
§
10'
ORAW
• PHIMJ
• ASECO
o
o o
•
" ° °° o 0Q° ° ° °0
• ° 0 m ° Qm mm 0 £
A " " B °"«"" • "• J 0,,* •" • "
• X A ^ ^ * •
^ B
AA
A
A
A
A
• A
„ | |A | | | | | | 1 1 1 1 1 1 J. 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 lA 1 1 J 1 1 1
APRIL JULY OCT |JAN APRIL JULY OCT JJAN APRIL JULY OCT JJAN APRIL
«Y
NDAMY
6. ~"
1
JULY 1
1
10s
-I01
10
FIGURE 3 Volatile suspended solids changes in raw-composite, primary and secondary aerated
lagoons
-------
,OT
io6
,o5
io4
10*
io1
0 RAW
• PRIMARY
0 A SECONDARY
0
ooo ooo o o
0
0 00
o o ooo o
o o o
• •" ••"•"" •"•• * .
000 •"•A • • • •
• A 4 •
• A O »
• A • A A
A A • A AA •• g -
A A * A A A
A A A A A
• A
A A A A A A
" * * A
A A
A A
1 1 1 1 1 !! 1 1 1 1 1 ! 1 I ! 1 1 1 1 1 1 1 1 1 ! 1 1 1 1 1 1 t I 1 1 1 U
APRIL JULY OCT |JAN APRIL JULY OCT IjAN APRIL JULY OCT IJAN APRIL
FIGURE 4 Coliform changes in raw-composite, primary and secondary aerated lagoons
ior
io6
.o5
io4
to
-------
.os
10*
o o o o
oo o o o o
o o
0° O
o 0
o • o
• •
•• gAA •
A A
A A
10"
I09
10'
A A
AA
.0'
10*
A A
I I I I I I I I I I I I I
I I
APRIL JU.Y OCT IJAN APRIL JULY OCT
•1966
196T
IJAN
$
APRIL JULY OCT IJAN APRIL
1968
1969-
FIGURE 5 Fecal coliform changes in raw-composite, primary and secondary aerated lagoons
-------
o
o
E
>*.
D
£
o
do
0 0°8
0 01 B° 0
0 • • ••
•••• • AAA ii 00
A A
A A
.o
10
io
APRIL JULY OCT
1966
j*"
APRIL JULY
1967 —
OCT
-H
JAN APRIL JULY OCT I JAN APRIL
1968 )k 1969
FIGURE 6 Enterococci changes in raw-composite, primary and secondary aerated lagoons
NJ
-------
IU
J
« I02
Q
O
CD
in1
m
•
A A
A *
A ft A A
i ^ A A
- * A. A* *^ 1
A A A
1 llliiiii A Aiiiiiit i
A COLIFORM -
* FECAL COLIFORM I
A* * A * ,
** A * A
A A I
A A A
A. A A
A A A -
A
A
1 T IIJIII 1 • •!••••
I01
I0
I0
10*
10
BACTERIA/100 ml
FIGURE 7 Relationship of BOD5 of secondary lagoon to coliform and fecal coliform organisms
-------
10
0
00
a
A
A
8
A
A
' I I > I Ilil 1 1 t I I I I I F|0
10'
I0
1
10'
ENTEROCOCCI/IOOml
FIGURE 8 Relationship of BOD5 of secondary lagoon to enterococci organisms
CO
o
-------
o
m
10"
I02
.0'
1C
- -•'.•"...
M ' - ' '
"•
••
: ' 4
; A * *
A
£ A A
£ A A^
A A
)4 IO5 IO6
• PRIMARY I
A SECONDARY -
1 ! : , • " '
A fe * • AB • -102
5 A .A -A A " . I
A* A •
I J A • *
A * A
A .ill U_L« 1 1 1 — I I I I t IO
IO7 IO8
BACTERIA/ml
FIGURE 9 Relationship of BOD5 of primary and secondary lagoon to total bacterial population
-------
303
APPENDIX
-------
304
TABLE 5
BODS Changes in Aerated Lagoons
BOD 5 mg/liter
Date Raw Primary Secondary
15 Nov 65 25
14 Dec 65 154 16
11 Jan 66 294 104 22
1 Feb 66 692 167 44
8 Feb 66 150 146 48
15 Mar 66 49 52
29 Mar 66 35 20
11 May 66 104 29
6 Jun 66 102 13
22Jun66 476 141 27
18Jul66 72 8
10Aug66 75 18
21 Sep 66 335 155 95
2 Nov 66 275 103 32
7 Dec 66 436 177 40
HJan67 309 209 69
9 Mar 67 262 219 96
22 Mar 67 308 172 57
19 Apr 67 309 146 55
23Aug67 263 77 30
6 Sep 67 198 97 26
20 Sep 67 221 76 18
3 Oct 67 245 90 17
18 Oct 67 165 69 23
1 Nov 67 240 96 23
15 Nov 67 440 102 24
29 Nov 67 295 96 29
13 Dec 67 320 181 73
27 Dec 67 2600 193 109
10 Jan 68 240 215 154
24 Jan 68 310 222 137
7 Feb 68 360 207 180
20 Feb 68 345 350 178
5 Mar 68 210 150
12 Mar 68 140
27 Mar 68 84 154 128
10 Apr 68 62 130 71
24 Apr 68 56 129 74
8 May 68 53 110 60
20 May 68 171 87 40
5 Jun 68 119 112 25
19 Jun 68 348 89 12
30 Jul 68 90 17
27Aug68 70 24
2 Oct 68 82 37
6 Nov 68 400 19
9 Dec 68 52 20
14 Jan 69 47 24
21 Feb 69 167
27 Mar 69 48 39
9 Apr 69 27 33
2 May 69 31 8
5 Jun 69 58 56
-------
305
TABLE 6
Total Nitrogen Changes in Aerated Lagoons
Nitrogen mg/liter
Date Raw Primary Secondary
14 Dec 65 14 7 50
11 Jan 66 17.0 12.6 69
1 Feb 66 29.2 13.7 7.4
8Feb66 12.8 10.9 6.9
15 Mar 66 46 53
29 Mar 66 6.9 3.8
11 May 66 11.8 4.0
6 Jun 66 9.1 4 6
22 Jun 66 24.3 10.3 5.5
18 Jul 66 2.9 3.4
10Aug66 4.0 3.8
21Sep66 14.1 17.5 11.8
2Nov66 15.4 13.7 9.5
7 Dec 66 24.7 17.5 6.7
11 Jan 67 14.8 16.3 9.9
9 Mar 67 16.7 12.0 12.2
22 Mar 67 18.6 12.0 12.2
19 Apr 67 17.3 10.3 8.4
23Aug67 13.1 11.2 7.6
6Sep67 12.7 12.7 7.2
20Sep67 11.8 10.3 4.9
3Oct67 13.9 11.8 5.9
18Oct67 11.8 8.7 5.3
lNov67 11.6 13.3 5.3
15Nov67 19.8 14.1 5.5
29Nov67 20.7 13.1 6.3
13 Dec 67 16.7 14.3 8.7
27 Dec 67 169.0 15.0 12.2
10 Jan 68 15.0 14.3 16.2
24 Jan 68 19.6 12.5 12.5
7 Feb 68 18.4 12.7 12.4
20 Feb 68 22.2 13.6 13.6
5 Mar 68 11.6 12.2
12 Mar 68 13.8
27 Mar 68 6.8 9.6 9.6
10 Apr 68 7.2 12.8 9.0
24 Apr 68 5.8 11.0 9.2
8 May 68 6.6 8.8 7.8
20 May 68 15.8 8.8 8.2
5 Jun 68 12.0 10.2 5.4
19 Jun 68 17.2 9.8 6.0
30 Jul 68 11.2 5.6
27Aug68 14.8 9.2
2 Oct 68 9-4 9.2
6Nov68 12.0 5.8
9 Dec 68 9.6 4.8
14 Jan 69 9.2 5.8
21 Feb 69 13.6
27 Mar 69 7.6 6.6
9 Apr 69 6-4 6.4
2 May 69 7-0 3.0
5 Jun 69 14.2 10.0
-------
306
TABLE 7
pH Changes in Aerated Lagoons
PH
Date Raw Primary Secondary
15Nov65 7.9
14 Dec 65 7.7 7.7
llJan66 8.4 7.6 7.7
!Feb66 7.5 7.5 7.7
8Feb66 8.3 7.7 7.6
15 Mar 66 7.4 7.7
29 Mar 66 7.4 7.4
11 May 66 7.8 8.0
6 Jun 66 7.9 8.1
22Jun66 7.4 7.9 8.1
18Jul66 7.8 7.8
10Aug66 7.7 8.0
2lSep66 7.8 8.0 8.1
2Nov66 8.1 7.7 7.9
7 Dec 66 8.5 7.6 7.6
llJan67 8.2 7.7 7.7
9 Mar 67 8.5 7.5 7.5
22 Mar 67 8.2 7.5 7.4
19 Apr 67 8.3 7.5 7.6
23Aug67 7.9 8.2 8.4
6Sep67 8.0 8.1 8.2
20Sep67 8.1 8.0 8.1
3Oct67 7.5 7.7 7.9
18Oct67 8.0 7.8 7.9
!Nov67 7.5 7.7 8.0
15Nov67 7.4 7.8 7.8
29Nov67 7.7 7.7 7.7
13 Dec 67 8.2 7.7 7.6
27 Dec 67 6.9 7.6 7.8
10 Jan 68 7.5 7.4 7.6
24 Jan 68 7.9 7.5 7.6
7 Feb 68 7.6 7.6 7.6
20Feb68 7.7 7.2 7.5
5 Mar 68 7.5 7.6
12 Mar 68 8.4
27 Mar 68 7.6 7.6 7.6
10 Apr 68 7.7 7.7 7.7
24 Apr 68 7.9 7.5 7.7
8 May 68 7.6 7.8 7.9
20 May 68 8.2 7.8 7.9
5 Jun 68 8.7 8.0 8.0
19 Jun 68 7.4 8.0 8.0
30Jul68 7.8 7.9
27Aug68 7.6 7.9
2Oct68 8.4 8.7
6Nov68 7.7 8.0
9 Dec 68 7.6 7.8
14 Jan 69 7.4 7.5
27 Jan 69 7.4 7.5
9 Apr 69 7.3 7.4
2 May 69 7.6 8.0
5 Jun 69 7.5 8 0
-------
307
TABLE 8
Solids Changes in Aerated Lagoons
Solids mg/Hter
Raw Primary Secondary
Date TVS VSS TVS VSS TVS VSS
15Nov65 301 24
14 Dec 65 436 144 294 36
11 Jan 66 536 220 274 116 363 96
1 Feb 66 676 316 376 148 346 104
8 Feb 66 386 152 418 132 368 52
15 Mar 66 258 80 261 92
29 Mar 66 250 16 184 0
11 May 66 349 92 293 40
6Jwi66 330 60 273 0
22Jun66 766 312 461 156 352 100
18 Jiil 66 582 92 530 84
10AUB66 528 216 588 160
2lSep66 450 136 396 124 400 60
2 Nov 66 444 140 316 120 320 72
7 Dec 66 688 224 450 112 295 30
11 Jan 67 470 184 392 172 332 80
9 Mar 67 480 236 488 112 312 72
22 Mar 67 512 228 278 64 294 68
19 Apr 67 558 192 276 98 235 52
23Aug67 490 114 582 160 498 70
6Sep67 388 122 474 104 446 46
20Sep67 460 200 462 100 486 46
30ct67 406 184 420 116 348 36
180ct67 476 212 418 104 426 36
1 Nov 67 438 164 332 124 346 48
15 Nov 67 590 220 312 100 278 20
29 Nov 67 496 144 268 92 292 24
i^TWfi? 590 204 468 160 o
-------
308
TABLE 9
Coliform Changes in Aerated Lagoons
Coliform/100 ml
Date
29 Mar 66
11 May 66
6 Jun 66
22 Jun 66
21 Sep 66
2 Nov 66
7 Dec 66
11 Jan 67
9 Mar 67
22 Mar 67
19 Apr 67
23 Aug67
6 Sep 67
20 Sep 67
3 Oct67
18 Oct 67
1 Nov 67
15 Nov 67
29 Nov 67
13 Dec 67
27 Dec 67
10 Jan 68
24 Jan 68
7 Feb 68
20 Feb 68
5 Mar 68
12 Mar 68
27 Mar 68
10 Apr 68
24 Apr 68
8 May 68
20 May 68
5 Jun 68
19 Jun 68
30 Jul 68
27 Aug 68
2 Oct 68
6 Nov 68
9 Dec 68
14 Jan 69
21 Feb 69
27 Mar 69
9 Apr 69
2 May 69
5 Jun 69
Raw
4x10*
5x10*
3x10*
2x10*
3x10?
5 x 10*
5 x 10*
5x10*
3x10*
1x10*
1x10*
1x10*
6 x 10*
5x10*
8x10*
3x10,
4x10*
1x10*
Ix
2 x 10,
8x10*
2x10*
3x10,
2x10*
9x10*
6x10*
1x10*
4x10*
3 x 10*
*
3xl06
Primary
4X10*
2x10°
6x10*
8x10*
7x10*
8x10*
1x10*
1x10*
3 x 10*
4x10*
2x10*
2x10*
7x10*
7x10*
5x10*
6x10*
9 x 10*
ixio*
5 x 10*
6x10*
ixio*
8x10*
9x10*
9x10*
4x10*
8 x 10*
7x10*
8x10*
6x10*
7x10*
5x10*
9x10*
1 x 10*
6x10*
5x10*
4x10*
9x10*
1x10*
1 x 10*
6x10?
6 x 10*
3x10*
9xlOs
Ix
1 x
Secondary
1x10*
2x10*
1x10!
1x10*
2x10*
6x10*
7x10*
10*
10*
2x10*
7x10*
2x10*
3x10*
1x10*
6x10*
2x10*
5x10*
8x10*
4x10*
2x10*
5x10*
1x10*
4x10*
6x10*
4x10*
9x10*
3x10*
1x10*
2x10*
1x10*
1x10*
5x10*
4x10*
2x10*
4x10*
6x10*
2x10*
4 x 10?
3x10*
2x10*
4 x 10*
3x10*
IxlO4
-------
309
TABLE 10
Fecal Coliform Changes in Aerated Lagoons
Fecal Coliform/ 100 ml
Date Raw Primary Secondary
29 Mar 66 1 * «* 4 x 10*
11 May 66 7 x 10s 2 x 10
6Jun66 ' _ 7x10 7x10
lxl° l^i x :
8xio< is1* n :
23*$ xl°o* 111%
r 3xlO« 3xlO« 2x10*
9Mar67 2 x 10' 1x10* 4x10
22 Mar 67 6 x 10s 1 x IjJ 5x10
19 Apr 67 1x10^ 3 x 10j 1x10
23Aug67 Ixioj 4x10 3x10
6Sep67 IxlQj 2x10 6x10
20Sep67 2xl0; 2x10 4x10
30ct67 3x10^ 2x10 3x10
180ct67 2x10^ 4x10 6x10
lNov67 lxl°7 oX^ fixlo4
15Nov67 3x10^ 9x10 6x10
29Nov67 5x10^ 1x10 9x10
13Dec67 2x10^ 2x10 6x10
27 Dec 67 2 x 10 J 3x10 2x10
10Jan68 1 x loj 3x10 2 x 10fi
?-s "«: :
12 Mar 68 3 x 10 6 8 x 10s
2?Mar68 4 x 10* J x 10 « x JJ5
10 Apr 68 5x10^ | x 10, J « JJ,
24 Apr 68 7x10 ^ x o.u s
8May68 1 x 10j 2x10 2 x 10g
20 May 68 3 x ICT 3 « 10ft 5 x 10,
5Jun68 4x10 2 x 10ft s
19Jun68 1x10 3 x 10fi 4
30 Jul 68 I x JJ6 2 x 10s
27Aug68 3xl05 4xl04
2 Oct 68 J J }J6 5 x 104
6Nov68 **{QS 8X104
9 Dec 68 Jx JJs 5 x 104
14 Jan 69 , « B X 1U
2lFeb69 4x10 5 3 x 104
27 Mar 69 * s ! x io5
X
9 Apr 69 QS 2 x IO4
2 May 69 2 x 10s 1 x IO3
5 Jun 69 2 X 1U
-------
310
TABLE 11
Enterococci Changes in Aerated Lagoons
Enterococci/100 ml
Date
11 Jan 66
1 Feb 66
8 Feb 66
15 Mar 66
29 Mar 66
11 May 66
6 Jun66
22 Jun 66
18 Jul 66
10 Aug 66
21 Sep 66
2 Nov 66
7 Dec 66
11 Jan 67
9 Mar 67
22 Mar 67
19 Apr 67
23 Aug 67
6 Sep 67
20 Sep 67
3Oct67
18 Oct 67
1 Nov 67
15 Nov 67
29 Nov 67
13 Dec 67
27 Dec 67
10 Jan 68
24 Jan 68
7 Feb 68
20 Feb 68
5 Mar 68
12 Mar 68
27 Mar 68
10 Apr 68
24 Apr 68
8 May 68
20 May 68
5 Jun 68
19 Jun 68
30 Jul 68
27 Aug 68
2 Oct 68
6 Nov 68
9 Dec 68
14 Jan 69
21 Feb 69
27 Mar 69
9 Apr 69
2 May 69
5 Jun 69
Raw
4 xlO4
2xl04
IxlO2
7xl04
8 xlO5
8xl05
8xl05
9xl05
2xl06
IxlO6
8xl05
7xlOs
9xlOs
6xlOs
3xl05
8xlOs
9xlOs
4xlOs
6xlOs
7xlOS
6xl05
5xl06
IxlO6
4 xlO5
8xl05
IxlO6
4xl05
2xlOs
2xlOs
8xlOs
4x10*
IxlO6
2xl05
2xl05
1x10°
Primary
IxlO4
4xl02
5xl02
IxlO5
8xl04
3xlOs
2xlOs
2xlOs
IxlO5
IxlO5
2xlOs
2xlOs
6xl05
4xl05
5xl05
4xlOs
2xlOs
4xl04
IxlO5
1x10
IxlO5
2xl05
2xlOs
2xl05
2xl05
7xl05
3xl05
2xl05
4xl05
4xlOs
3xl05
4xl05
2xlOs
2xl05
3xlOs
3xlOs
IxlO5
IxlO5
IxlO5
IxlO5
9xl04
IxlO4
5xl04
2 xlO4
4xl04
6xl04
5xl04
8xl04
3xl04
Secondary
9xl03
4 xlO2
IxlO2
3xl03
IxlO4
IxlO3
IxlO2
IxlO2
IxlO2
6xl02
3xl03
6xl03
7xl04
2xl05
2xl05
4xl04
2 xlO3
3 xlO3
8xl02
2xl03
2 xlO3
8xl03
7 xlO3
2 xlO4
IxlO5
6xl04
9 xlO4
IxlO5
2xlOs
2xl05
2 xlO5
1 x 10s
8 xlO4
7 xlO4
4 xlO4
IxlO4
3 xlO3
2 xlO3
IxlO3
2xl03
IxlO3
3xl03
4 xlO3
5xl03
3xl03
IxlO4
2 xlO3
-------
311
TABLE 12
Total Bacterial Changes in Aerated Lagoons
Bacteria/100 ml
Date Raw Primary Secondary
llJanSG 3x10* 2x10* 6x10*
!Feb66 7xl07 1x10° 5 x 10
8Feb66 4 x 107 8 x HT 1 x 10
15 Mar 66 8x10* 7x10
29 Mar 66 3 x 108 3 x 10* 8 x 10
11 May 66 4 x 10» 5 x I0j
6Jun66 g 9x10' 2 x 10 J
22Jun66 2 x 109 1 x 109 3x10
18 Jul 66 2 x 10* 4 x 10*
10Aug66 8 2x10*, 2x10
21Sep66 9x10* 1 x 1Q» 9x10
2Nov66 9x10* 4 x 109 3x10
7 Dec 66 5 x 108 6 x 10° 2 x 10
11 Jan 67 4x10* 3x10* 3x10
9Mar67 3 x loj 2 x 10* 3x10
-IS ' ?:X: x *
20 67 3x10^ 9x10 7x10
67 3x10 2x10* 2x10
*
180ct67 2xi09 5x10* 3x10
!Nov67 2xl09 1x10 2x10
15Nov67 SxlO9, 7x10 3x10
29Nov67 2xl09 3x10 2x10
13Dec67 1 x 109 2 x 10* 1 x 10g
27Dec67 9xl09 4x10 4x10
10 Jan 68 2 x 109 3 x 108 2x10*
^FebeS 3xl09 3xl09 2x10*
20Feb68 2 x 109 3x10 2x10
x
27Mar68 2 x lo 2x1 x
10 Apr 68 2x10^ 3x10 2x10
24APr68 2xl°9 iXJS« 6xio8
8 May 68 2x10^ 4x10 6x10
20 May 68 2 x 10* 1x10 1x10
5 Jun68 Gxioj 1x10 5xlO?
19Jun68 4xl09 2x10 8x10
qn Tlli CQ 1 x 10n d x 1U7
??A^f« 6xl09 9x10^
27 Aug 68 9 4x10*
4x10 *xiu
8
6Nov68 9 7
69 8
21Feb69 2x10 . 2x10*
27Mar69 4x10 2 x 108
9Apr69 3xl09 2 x 10g
-------
DISINFECTION AND TEMPERATURE INFLUENCES
Cecil Chambers and Gerald Berg
INTRODUCTION
The purpose of this discussion is to explore a number of disinfection methods, some conventional,
and some not so conventional, and to examine their potential strengths and weaknesses for
disinfecting effluents under cold climate conditions. A second objective will be to outline the need
for research to assist in developing guidelines for the most efficient practical use of disinfectants
where extremely low temperatures prevail.
Disinfection, like primary settling, aeration, and other steps in waste treatment, cannot be
considered independently of other phases of the process. All of these factors affect the quality of
the final effluent produced. This is especially important in the situation we are concerned with,
because the arctic environment influences the various methods of waste treatment in ways which
have profound effects on disinfection efficiency.
In general, effluents in arctic areas are relatively crude. A majority of the treatment plants in
Alaska have only primary settling. The high turbidity and organic content of such effluents make
them difficult to disinfect by any method available. In addition, because biological oxidation
proceeds at a slow pace at low temperatures, the effluents from secondary activated sludge plants
and other biological treatment processes are incompletely oxidized and high in ammonia nitrogen
content.
Because its primary purpose is the prevention of transmission of disease, disinfection is considered
the most critical step in waste treatment. Nevertheless, application of massive doses of chlorine,
for example, should not be accepted as a substitute for adequate removal of solids and
stabilization of soluble organic matter. When such crude effluents are discharged to receiving
waters, the low biological activity at arctic temperatures will contribute to an increasing and
cumulative load of oxidizable material, especially in lakes, streams, and estuarine waters.
Dissolved oxygen levels frequently reach extremely low concentrations under ice cover in Alaska.
Under these circumstances, any attempt to circumvent adequate treatment by excessive
application of disinfectants will further deplete the critically low dissolved oxygen supply. In
addition, the question is being asked with increasing frequency, are we, in applying massive doses
of chlorine to crude effluents, creating toxic chlorine addition products that adversely affect the
flora and fauna in the receiving water? What is the effect on the aerobic food chain? When such
possible toxicity is superimposed on the effect of an already precariously low oxygen level, it may
lead to the demise of a critical segment of the food chain. This can be especially important in cold
climate areas where biological productivity of waters is low. The consideration of non-residual
disinfectants may have potential to resolve or minimize this problem in some low temperature
situations. Another approach is neutralization of the disinfectant before discharge or selection of a
chemical disinfectant that is effective at minimum residual levels.
The efficiency of chemical disinfectants decreases with decreasing temperature {Clarke and Chang,
1959). Another problem that should be considered is the extended survival of organisms at
312
-------
313
near-freezing water temperatures. Both enteropathogenic and coliform bacteria are known to
survive in water for extended periods of time at very low temperatures, and viruses survive for long
periods when frozen in ice. Clark (1970) has pointed out that in Alaska, "Stream and lake waters
will be used for drinking water for some time to come without treatment in spite of all efforts to
provide, require, and encourage treatment."
Because of prolonged survival of pathogenic organisms in cold water and ice, the organisms may be
transported by currents under the ice and by drifting ice when break-up time comes to serve as
sources of infection far from their point of origin. In addition, there is widespread groundwater
contamination in many areas especially during spring thaws. This is further aggravated by
inadequate disinfection of wastes and extended survival of contaminants at low groundwater
temperatures which are frequently just above the freezing point. These conditions combined with
the widespread practice of drinking untreated water emphasize the importance of thorough
disinfection of effluents in cold climate areas.
GENERAL AND BACTERICIDAL CONSIDERATIONS
Chlorine
Chlorine is more widely used for disinfection of water and wastewater than any other chemical
agent. Regardless of whether chlorine is added to water as liquid chlorine (CI2), sodium
hypochlorite (NaOCI), or calcium hypochlorite {Ca(OCI)2}, the same disinfecting entities will be
produced (Fair et al., 1948). In wastewater disinfection, two forms of chlorine are of primary
interest: they are (1) hypochlorous acid (HOCI) and (2) chloramines.
A third form, hypochlorite ion (OCI~), has little if any disinfecting efficiency. The hydrolysis
product of the reaction of chlorine and water, HOCI, is the predominant form in which chlorine
exists in aqueous solution between pH levels of 2.0 and 7.0. Between pH 7.0 and 8.0 progressive
ionization of the HOCI takes place. At the higher pH level most of the chlorine exist as OCI' ions.
Significant amounts of ammonia-nitrogen, probably 20 ppm or more, can be anticipated in arctic
effluents. This ammonia will combine with chlorine to produce chloramines of which
monochloramine is the dominant, and probably the mostgermicidal, form of chlorine that will be
present at the pH of most effluents (Baker, 1959). The exception would be break-point
chlorination, where ammonia-nitrogen is destroyed by application of chlorine at a
chlorine-to-nitrogen ratio of approximately 9 to 1 (Butterfield, 1948a) to provide residual HOCI.
Chlorine demand, caused by organic components of sewage, is a serious problem in disinfection of
wastewater. Assuming no chlorine demand, approximately 180 ppm of titratable chlorine would
be required to reach the breakpoint and provide residual HOCI at an ammonia-nitrogen
concentration of 20 ppm. Except initially, HOCI is not likely to be present in most chlorinated
effluents.
Monochloramine is a relatively inefficient disinfectant when compared to HOCI (Butterfield,
1948b). Butterfield and Wattie (1946) in tests with chloramine* using Escherichia coli, Salmonella
*While the term chloramine is used in their report, conditions were such that monochloramine was
the dominant form produced.
-------
314
typhosa. and Shigella sonnet as test organisms, concluded that, "A reduction of 20° C in
temperature (20° C - 25° C to 2° C - 6° C) requires 9 times the exposure period, or 2.5 times as
much chloramine to produce a 100% kill." Decreasing the hydrogen ion concentration decreases
the disinfecting efficiency of both HOC! and monochloramine.
When the effects of an unfavorable temperature and pH are combined, the disinfecting efficiency
of monochloramine is very seriously depressed; these effects are demonstrated by Butterfield's
(1948fa) data which are presented in Figure 1. The advantages of maintaining the lowest possible
pH, consistent with avoiding corrosion problems, are readily apparent when using HOC! or
monochloramine as a disinfectant at low temperature.
Iodine
At acid pH levels, iodine exists primarily in the elemental state and forms tri-odide and higher
iodides in the presence of increasing quantities of iodide ions. At increasing pH levels, elemental
iodine hydrolyzes to hypoiodous acid (HOI) which, in the absence of excessive iodide ions, may
constitute about 40% of the iodine at pH 8.0. As the pH increases beyond this level, the HOI
decomposes to iodates which are essentially non-germicidal. Elemental iodine and several of its
derivatives are efficient disinfectants, although not quite as efficient as HOCI. While iodine is more
expensive than chlorine, it does not form iodamines and has relatively low reactivity to organic
matter (Chang, 1966). Accordingly, iodine has considerable potential for circumventing some of
the problems encountered with chlorine.
CO
Ul
a:
AFREE (HJDWNE
at 20iwn -
aoe
9 10 II
pH
100% KILL OF ESCHERICHIA COL1
FIGURE 1 The combined effort of variations in pH and temperature on the disinfecting
efficiency of free chlorine and chloramine. After Butterfield (1948b)
-------
315
P-6 Sityphl-murium
C-36 E. coll
P-9 S sonnel
FIGURE 2 Average ppm iodine required to kill all test bacteria in 1 minute. After Chambers et
al. (1952)
Chambers et al. (1952) in tests with E. coli, Aerobacter aerogenes. Streptococcus fecdis, and three
species of Salmonellae, including S. typhosa, and four species of Shigellae, concluded that, "At a
given exposure time and pH the iodine concentration required to kill at 2° C to 5° C may be as
much as four times greater than at 20° C to 26° C," and ". . . increases in pH reduce the
bactericidal action of iodine." These effects are illustrated in Figure 2 which presents the results
obtained with a pollution index organism, E. coli, and the two most resistant enteropathogenic
bacteria tested, Salmonella typhimurium and S. sonnet. Recent information indicates that the
most desirable pH range for disinfection with iodine is pH 7.0 to 8.0 (Berg, 1970).
Because iodine has not been widely used for disinfection of water and wastewater, methods of
application are not as universally understood as is the case with chlorine. Chang (1966) has
designed a system for application of iodine to water that can be adapted easily to disinfection of
effluents. His report also presents an excellent bibliography on disinfection of water with iodine.
In addition, a patent has been granted on a system designed to convert iodine to the gaseous state
for use in disinfection of water (Starbuck, 1967).*
Excess Lime
In water and wastewater treatment, lime is usually considered primarily as a flocculating agent.
While there are some problems such as sludge disposal associated with the use of lime, lime
flocculation can markedly reduce the solids content of raw sewage and the overall BOD reduction
may be as much as 70% (Bishop and Sanoworth, 1970).
*References to commercial products are not to be construed as endorsement by the Federal Water
Quality Administration.
-------
316
Flocculation with lime can remove or destroy a relatively high percentage of the microorganisms
present. What is not generally realized is that lime can also be used as a disinfectant if pH and
contact time are maintained at suitable levels, and it is not subject to the organic demand problems
of halogen disinfectants and ozone. The data of Wattie and Chambers (1942), presented in Figure
3 show the disinfecting efficiency of increasingly high pH at temperatures of 0° C to 1° C and 20°
C to 25° C with E. coli.
Results obtained by the same people in comparable tests with S. typhosa are presented in Figure 4.
Riehl et al. (1952) augmented this work in subsequent studies with a wider spectrum of
pathogenic organisms at several temperatures ranging from 2° C to 25° C. Their results are
comparable to those presented in Figures 3 and 4.
Ozone
Although ozone is widely used as a disinfectant throughout Europe, its primary use for treating
drinking water in the United States is for the removal of taste, odor, and color.
There is very little information in the literature relating to the possibility of using ozone as a
disinfectant for waste treatment plant effluents. Stumm (1958), however, stated ". . . ozone has to
be considered as a potential disinfectant for water and sewage." He also indicated the need for
research in his concluding statement ". . . more investigations are needed to evaluate the
applicability of ozone as a practical tool for the sanitary engineer in water treatment and waste
disposal." Recently Smith (1967) presented data obtained in studies with effluent from a
municipal activated sludge plant. Tests were run with non-nitrified effluent, nitrified effluent, raw
g
t'""**^."
x o •"•£•;.
90
80
TO
60 j \
I
, 501
|
4O
I
30
20
W \ \
DH 1001-1050 ,.'• \
20-25° C +. ^
W)
v ,nH ii a-11 so
\ O-I'C
. .pHlLOI-1150 ''•;
I0|" V 20-25°C
••-. v?
pH 9.5I-IO.OO
0-PC
••v ..........
\ W
\I)H QJ-1050
0-1° C
\
W
* ,PH 951-10.00
•. \ 20-25° C
23456789
TIME IN HOURS
10
FIGURE 3 Death rate of Escherichia coli at various pH ranges at 0° - 1.0° and 20° - 25° C. After
Wattie and Chambers (1942)
-------
317
100
90
8O
_j
^ 70
>
OL
ID
CO 60
UJ
1*
1 4°
a
3O
20
10
0
<
^^i.
r '• * ^ *v
1- • s ^^
T ' \ ^
\- . x \ V
i '. \ \
"'• \ \
\ '• * v v
. \ . \ \
i • • N o \
'. v \
\ • '-. \ \
> l \ : \ \
i • % \ \
1 \ * »'•-. \
' \ \o \
' ,0 '^j. O\ {• °i \
' \ & ?,. V DH^I IN
1 -^* „ L. \ to 1000
1 \ '•.«» '-. x \ o
^ •*) . "*• \^>
- , (+ x) • . ~&>
'.(noKV -1". ( pH 10.01 ',-. ^
.\ PH ^ •. to 10.50 '•• * X?1
\toil50\ '"•-. ' .^
• *%> %> " '* *
X^y -^ft '"•-..
, v'* ^ ' >-., ^ T '
31 3456
' ' • •
-
-
.
-
_
?/.
>4*»-_»_
-
v x
**' K It'
' -f- .."t.^*.- _1 1
7 8 9 10
TIME IN HOURS
FIGURE 4 Death rate of Salmonella typhosa at various pH ranges at 0° - 1° and 20° - 25° C.
After Wattie and Chambers (1942)
sewage filtered through glass wool, and raw sewage (aluminum sulfate precipitated). They
concluded that disinfection at the 99% kill level was not attained with secondary effluents, and
results with raw sewage were less acceptable. In subsequent laboratory pilot scale studies they
concluded that ". . . an ozone dose of 7 to 10 ppm may be sufficient for disinfection of municipal
sewage treatment plant effluent under these circumstances."
In considering the possible applicability of ozone for disinfection of wastes under cold climate
conditions, the fact that ozone is a very powerful oxidizing agent makes it susceptible to organic
demand. This characteristic is offset to a degree by the fact that some reduction in BOD occurs
and the end products of the oxidation are non-toxic. In addition, ozone is rapidly dissipated and
no toxic residual disinfectant remains in the effluent. A distinct advantage is the fact that ozone is
not susceptible to reductions in efficiency in the presence of ammonia, as is the case with chlorine.
Increased solubility at the low temperatures encountered in the Arctic may also be an advantage.
However, the excessively high iron content of some Alaskan groundwaters (6 to 14 mg/l common;
Clark, 1970) may be a problem if the iron is in the reduced state and such waters find their way
into the disposal system.
Bromine
Bromine is recognized as a powerful bactericidal agent. Although it is somewhat more expensive
than chlorine, bromine has been used with considerable success as a swimming pool disinfectant in
Illinois (Klassen and Sieg, 1948). Nevertheless, until recently, little attention has been given to
investigation of bromine for use in disinfection of water and even less consideration has been given
to the potential of bromine for disinfection of wastewater. The work of Johannesson (1960) has
-------
318
provided a basis for a better understanding of the fundamental behavior of bromine as a water
disinfectant, and this has served as a stimulus to increased interest in bromine. Because
bromamines are essentially as germicidal as free bromine, a highly potent disinfectant, bromine
may have potential for use in disinfection of waste treatment plant effluents especially in cold
climate areas where effluents with high ammonia content may be expected to be a continuing
problem. Bromine has the disadvantage, however, of being relatively susceptible to organic demand
and elevated pH (9.0-9.5) is necessary to produce significant amounts of bromamines.
The Federal Water Quality Administration has been sponsoring research to determine the relative
efficiency of chlorine and bromine for disinfection of activated sludge plant effluent (Solloetal.,
1970). In this work, both dosage and residuals of chlorine and bromine are being correlated with
percent survival of coliforms, fecal coliforms, and total bacterial numbers. Anomalous results have
been obtained with chlorine at 30° C (pH 9.0) and these results cannot be explained. Control of
bromine concentrations has proven difficult because of the rapid bromine decay rate.
Combinations of chlorine and bromine are also being evaluated, and it appears that pretreatment
with chlorine, followed by bromine, may produce effective disinfection at a lower cost than
bromine alone. The preliminary results obtained indicate that ". . . the efficiency of disinfection is
almost independent of the chlorine dosage, so long as it is greater than the immediate demand.
Thus, the function of the clorine appears to be only that of conserving bromine by satisfying
demand." The minimum test temperatures in this project were 10° C, but this work is being
continued and tests at 0° C to 2° C are planned for the future.
Pasteurization
Pasteurization, because it is unaffected by the chemical quality of wastes, may be feasible in arctic
climates in some locations. Sludges and effluents from sanitoria have been disinfected with heat in
a number of European countries and sewage sludge has been pasteurized before spraying on
agricultural land in Germany (Kugel, 1968). Goldstein et al. (1960) tested a system for
pasteurization of water that was effective in destroying coliform organisms. They considered the
principal advantages to be reliability and simplicity, both important factors in Alaska. On a
household scale, costs were estimated to be approximately $1.00 per 1,000 gallons, but they
would have been lower if the unit had been operated constantly. At cold climate temperatures,
costs would be somewhat higher and revised design criteria would probably be necessary with
wastewater to minimize fouling of heat exchangers. Construction plans for this unit have been
published (U.S. Public Health Service, 1959). Seiberling and Harper (1955) have also reported on a
pasteurizer design for successful high temperature short time exposure of water. Harper stated
(personal communication to C. W. Chambers, 1965) that after approximately 10 years of
operation ". .. this unit has been extremely satisfactory..."
Miscellaneous Disinfectants
Gamma radiation has been advocated by some people for disinfection of wastes. The fact that it is
a physical process might indicate potential for low temperature use, but there appears to be little
in the work reported to offer encouragement. There have been reports of a marked synergistic
action of gamma radiation on the disinfecting action of chlorine, and this could be advantageous
under cold climate conditions; however, the work reported has been inadequately con trolled. The
authors are investigating this phenomenon and results to date show no evidence of any practically
significant synergistic effect with monochloramine when conditions are adequately controlled.
-------
319
Ultraviolet light has some potential for disinfection of high quality effluents. The types of
effluents produced in Alaska may be high in iron and color content; and both of these interfere
seriously with the disinfecting efficiency of ultraviolet light. Another trait that has to be
considered is the fact that low ambient temperatures reduce lamp efficiency. It is possible that
ultraviolet light may have some potential for use with effluents that have been highly clarified by
chemical coagulation, provided interference from color and residual lime floe are not encountered.
o
o
.1
.0*
.08
.07
.06
.09
I I I I I
A .6 .7JB.»I
2 3 4
MINUTES
667S9IO
FIGURE 5 Inactivation of poliovirus 1 by HOCI. After Weidenkopf (1958)
VIRUCIDAL CONSIDERATIONS
Chlorine
Hypochlorous acid (HOCI) is a rapid virucide (Fig. 5) (Weidenkopf, 1958), while the OCI" ion is a
poor virucide, if in fact it kills any viruses at all. In a clear aqueous solution, even at low
temperature, chlorine is a very rapid virucide at acid pH, and a very slow virucide beyond mildly
alkaline pH levels. Chloramines are slow virucides (Fig. 6) (Shuval, 1966), usually too slow for
effective treatment of virus-bearing water. Even the quick virucidal activity that occurs
immediately after chlorine is added to a sewage effluent, an activity that reflects the action of
HOCI before it has all reacted to form chloramines, leaves a large amount of virus to the
slow-killing chloramines. Since even very small numbers of viruses are capable of producing
infections (Plotkin and Katz, 1967), it is clearly important that all viruses should be removed from
-------
320
any waters with which man may come in contact. Thus, when HOCI cannot be maintained, or
when its presence cannot be tolerated, we must consider other disinfectants.
Survival of Viruses in Sewage and Water
Freezing brings about the formation of ice crystals in bacteria, and with this, eventual disruption
of the cellular membrane of many cells. The result is a gradual to rapid die-off of bacteria, the rate
depending on the species, temperature, and other environmental factors. Viruses are not cellular,
and freezing does not destroy them. To the contrary, freezing is the best known method for
storing viruses, and the lower the temperature, the better. In the laboratory viruses are preserved
by cold-temperature storage. Viruses may be stored at -70° C for a decade or more. Thus, in the
cold environment of the North, the normal die-off of viruses experienced in warmer climates
probably does not occur.
Figure 7 (Berg, 1966) shows the relationship between temperature and time for the destruction of
99.9% of three different viruses in stored sewage. This graph was prepared from data obtained by
others (Clarke, Berg et al., 1962). At 10° C, 99.9% destruction of echovirus 7 took more than 100
days, and both polivirus 1 and echovirus 12 required more than 60 days. Thus, enteroviruses can
survive in sewage for many months at low temperatures. In ice, survival time will exceed cold
water survivals and, of course, the lower the temperature, the longer will be the survival.
8.5mg/l residual (combined) chlorine
-II mg/l applied chlorine
Shuval *t al (1966)
____Lothrup and Sproul
(1969)
3 4
HOURS
FIGURE 6 Inactivation of poliovirus 1 by applied chlorine
-------
321
ECHOVIRUS 12
ECHOVIRUS 7
POLIOVIRUS I
20
TEMPERATURE 'C
FIGURE 7 Relationship between time and temperature for 99.9% reduction of virus in stored
sewage. After Clarke, et al. (1962) as revised by Berg (1966)
Viruses do not survive as long in river water as they do in sewage (Clarke, Berg et al., 1962; Berg
and German, 1970) and there is some evidence that they do better in lightly polluted water than in
heavily polluted water (Clarke. Berg et al., 1962). Their survival in distilled water is also of long
duration.
There is no simple explanation for survival patterns of viruses in waters of various qualities. Since
autoclaving increases survival in certain river waters, microbial life, their enzymes or other
heat-labile matter may account for some of the pattern. Certain divalent cations are known to
increase the survival of some viruses, and decrease it for others. Clearly, the chemistry of the
water, as well as the temperature, is an important determinant of virus survival.
Iodine
Elemental iodine is a rapid virucide, but not as rapid as HOCI. Elemental iodine, however, can be
effective in circumstances that impede the effectiveness of HOCI. In the presence of ammonia, for
example, where HOCI reacts to form the relatively slow chloramines, the iodine remains free to
react with viruses. Elemental iodine is a faster virucide than chloramine (Berg and Berman, 1970).
-------
322
349 678910
20 30 401
MINUTES
FIGURE 8 Inactivation of coxsakievirus A9 by elemental iodine. After Berg (1964)
At pH 8.0, where most HOC) has ionized to the poorly virucidal OCI" ion, much of the elemental
iodine has hydrolyzed to HOI which is considerably more virucidal than the elemental iodine
(Berg, unpublished data).
Figure 8 shows the relationship between time and elemental iodine concentration for the
destruction of 99% of coxsackievirus A9 over a three temperature range (Berg, Chang and Harris,
1964). Four other viruses tested under identical conditions were only slightly less resistant. E. coli,
Pseudomonas aeruginosa, Alcaligenes fecalis and Staphylococcus aureus were about equally
sensitive to elemental iodine, and depending upon temperature, about 1,100 to 1,700 times more
sensitive than coxsackievirus A9 (Berg, unpublished data). The QIC for coxsackievirus A9 was
about 4.5, but it was apparently less for E. coli.
From Figure 8, it is possible to determine both the times for 99% destruction of the virus over a
continuous temperature range at constant I? concentrations, or alternately, I2 concentrations for
99% destruction of the virus over a continuous temperature range at constant times (Berg, 1970).
Ozone
Despite its long widespread use in Europe, there are relatively few data available on the virucidal
efficiency of ozone.
Figure 9, however, gives an approximation of what ozone can achieve. Poliovirus 1 was inactivated
at a rapid rate in river water with only a few tenths of a mg/l of ozone, apparently at about 25° C
-------
323
(Coin, Hannoun and Gomella, 1964). Poliovirus 2 and 3 were inactivated more quickly under
similar circumstances (Coin, Gamella et al., 1967).
10,000 -
o
n
16
IB
7 10 13
MINUTES
FIGURE 9 Inactivation of poliovirus 1 by ozone. After Coin, et al. (1964)
Ultraviolet Light
Ultraviolet light (UV), at a wavelength of about 250-260 m/a, is strongly bactericidal. Only limited
data exist, however, on the virucidal capability of UV light. In distilled water at intensities of
4,000-11,000 Mw-seconds per cm. Huff et al. (1965) inactivated at least 99% of polioviruses 1,2
and 3, coxsackievirus A9, and echovirus 7. Only at lower dosages did small amounts of viruses
survive. In the presence of 9 standard units of color, the UV was slightly less efficient. Hill,
Hamblet and Banton (1969) evaluated UV, presumedly in the 250-260 mn range, exposing
poliovirus 1 in seawater to 83/iw/cm2 of radiation 14 cm from the source. Although the dosage is
not completely clear, the rapid virucidal effect of the UV is evident (Fig. 10).
-------
324
Gamma Radiation
Gamma radiation rapidly destroys viruses. In a menstruum containing salts and serum 0.5
megarads destroyed 90% of poliovirus 3. In distilled water, where quenching of free radicals
produced by the radiation did not occur, 90% of the virus was destroyed by less than 0.11
megarads (Sullivan, 1970). Many other viruses tested fell into generally similar resistance patterns.
There is little reason to anticipate, however, that economically feasible methods of disinfecting
wastewater with gamma radiation will be developed.
.01
10 19 20 28 30
UV EXPOSURE (second*)
SO
FIGURE 10 Incativation of poliovirus 1 with UV light. After Hill, et al. (1969)
-------
325
RESEARCH NEEDED
Lotspeich (1969) states, "Design criteria are simply not available - and this includes Scandinavia
and Soviet Russia as well as Alaska - to build disposal systems that will function with assurance
under the severe climatic conditions of northern latitudes." To those of us interested in
disinfection, this indicates that research to solve the problem of effective removal of solids and
soluble organic material should have a very high priority if adequate disinfection by chemical
methods is to become a practical reality under cold climate conditions.
Information should also be secured to provide satisfactory answers to questions regarding the
survival of bacteria and viruses under natural conditions in water and bottom sediments under ice.
This information is badly needed to determine the effect of an influx of crude and inadequately
disinfected wastes on survival patterns under ice cover. Comparable studies of the same receiving
waters are needed during open water periods to determine the effect of some of the violent
seasonal variations that occur in the Arctic.
It is also well recognized that a significant die-off of coliform and pathogenic bacteria occurs in ice
while viruses survive for relatively long periods. What, however, is the effect on these survival
relationships when incompletely disinfected effluents are frozen, especially when contact time has
been inadequate and some residual disinfectant remains? Research is needed to provide answers to
these questions.
The disinfecting potential of lime should be thoroughly investigated. It may be possible to
combine successfully the use of lime as a disinfectant with its ability to physically remove
microorganisms and a high percentage of solids and oxidizable material. Such research should
begin in the laboratory, using raw and/or settled primary sewage at very low temperatures. The
effect of increasing pH should be related to the holding time required to yield adequate
disinfection under the respective test conditions.
Bench-scale pilot plant studies with lime, or other disinfectants considered subsequently, should be
initiated just as quickly as the preliminary research provides adequate guidelines.
Iodine should be investigated in parallel laboratory tests with chlorine using different types of
effluents. The inefficient disinfecting properties of chloramines, especially with viruses, combined
with the high chlorine demand of many effluents are recognized. These factors create a need for
heavy chlorine dosage if adequate disinfection is to be accomplished. The organic demand effects
of effluents on iodine are not known. Work is needed to evaluate these effects of effluents on the
disinfecting efficiency of iodine. The research outlined should provide a basis for establishing the
true comparative efficiency of iodine and chlorine for low temperature disinfection of wastewater.
Preliminary research on the use of ozone should be considered to determine whether ozone
demands, especially of the types of effluents to be treated in Alaska, would preclude its use. Some
guidelines regarding research with ozone for disinfection of individual household wastes (toilet
wastes only) may be available from Smith (1967) who mentions successful treatment of such
wastes over a six-month period.
It should be emphasized that disinfection research should get out of the laboratory and into the
field for practical evaluation just as quickly as possible. The author is aware of the limited research
-------
326
resources available in most cold climate areas. Nevertheless, wherever possible, it seems desirable to
have the laboratory research most closely related to practical application done in the general area
where the field evaluation will be made. In this way, personnel most familiar with the technical
problems encountered in the laboratory and bench-scale pilot application studies will be available
for consultation and assistance in expediting field use.
In conclusion, I should like to point out that all disinfectant research should be planned to provide
the best possible basis for determining overall costs for application in various locations. In Alaska,
for example, the cost of shipping lime would probably be prohibitive in some inland locations, and
ozone would require electrical service. We need to develop a research program to provide a
spectrum of proven disinfection processes from which a sanitary engineer can select the one most
applicable to a specific waste disinfection problem. Only through such an approach will we be able
to satisfy the wastewater disinfection needs of Alaska and similar cold climate areas.
REFERENCES
Baker, R. J. (1959) Types and significance of chlorine residuals, J. Amer. Water Works Assn.,
51:1185-1190.
Berg, G. (1966) Virus transmission by the water vehicle. Health Lab. Sci., 3:90-100.
Berg, G. (1970) Virus inactivation and removal, Presented at the National Specialty Conference on
Disinfection, University of Massachusetts, Amherst, Massachusetts.
Berg, G. and Berman, D. (1970) Unpublished data.
Berg, G., Chang, S. L. and Harris, E. K. (1964) Devitalization of microorganisms by iodine, 1.
Dynamics of the Devitalization of Enteroviruses by Elemental Iodine, Virology, 22:469481.
Bishop, D. F. and Sanoworth, R. B. (1970) Monthly Report FWQA Contract No. 14-12-818 with
the District of Columbia, Dept. of San. Engineering, Available from Advanced Waste
Treatment Research Laboratory, FWQA, U.S. Department of the Interior, Cincinnati, Ohio.
Butterfield, C. T. (1948a) Bactericidal properties of free and combined available chlorine, J. Amer.
Water Works Assn., 40:1305-1312.
Butterfield, C. T. (1948b) Bactericidal properties of chloramines and free chlorine in water. Pub.
Health Rept. (U.S.) 63:934-940.
Butterfield, C. T. and Wattie, E. (1946) Influence of pH and temperature on the survival of
conforms and enteric pathogens in water, Pub. Health Rept. (U.S.) 61:157-192.
Chambers, C. W., Kabler, P. W., Malaney, G. R. and Bryant, A. (1952) Iodine as a bactericide,
Soap and San. Chem., 28:149-165.
Chang, S. L. and del Aqua, Yodacion (1966) lodination of water, Boletin de la Oficina Sanitaria
Panamericana, 4:317-331. (Mimeograph, English, available from Dr. S. L. Chang, U.S. Public
Health Service, Bureau of Water Hygiene, Cincinnati, Ohio.)
-------
327
Clark, S. E. (1970) Personal communication, Alaska Water Laboratory, Fairbanks. Alaska.
Clarke, N. A. and Chang, S. L. (1959) Enteric viruses in water, J. Amer. Water Works Assn
51:1299-1317.
Clarke, N. A., Berg. G., Kabler. P. W. and Chang, S. L. (1964) Human enteric viruses in water:
source, survival and removability, International Conference on Water Pollution Research
London (1962), Pergamon Press, New York, N. Y. (523-542).
Coin, L, Hamnoun, C. and Cornelia, C. (1964) Inactivation by ozone of poliomyelitis virus
present in water, La Presse Medicale, 72:2153-2155.
Coin, L., Gomelta, C., Hamnoun, C. and Trimoreau, J-C (1967) Inactivation by ozone of
poliomyelitis virus present in water (Further contribution). La Presse Medicale
75:1883-1884.
Fair, G. M., Morris, J. C., Chang, S. L., Weil. I. and Burden, R. P. (1948) The behavior of chlorine
as a water disinfectant, J. Amer Water Works Assn., 40:1051-1061.
Goldstein, M., McCabe, L. J., Jr. and Woodward, R. L. (1960) Continuous-flow water pasteurizer
for small supplies, J. Amer. Water Works Assn., 52:247-254.
Harper, W. J. (1965) Personal communication, Ohio State Univ., Department of Dairy Technol.,
Columbus, Ohio.
Hill, W. F., Jr., Hamblet, F. E. and Benton, W. H. (1969) Inactivation of poliovirus type 1 by the
Kelly-Purdy ultraviolet seawater treatment unit, Appl. Microbiol., 17:1-6.
Huff, C. B., Smith, H. F., Boring, W. D. and Clarke, N. A. (1965) Study of ultraviolet disinfection
of water and factors in treatment efficiency. Pub. Health Rept. (U.S.) 80:695-705.
Johannesson, J. K. (1960) The bromination of swimming pools. Am. J. Pub. Health,
50:1731-1736.
Klassen, C. W. and Sieg, J. G. (1948) Swimming pool operations. State of Illinois Department of
Public Health Circular No. 125.
Kugel, G. (1968) Jahresbericht of the Niersverband Gruppenklaranlage, Viersen (near Cologne,
Germany).
Lotspeich, F. B. (1969) Water pollution in Alaska: present and future, Science, 166:1239-1245.
Plotkin, S. A. and Katz, M. (1967) Minimal infectious doses of viruses for man by the oral route,
Transmission of Viruses by the Water Route (G. Berg, ed.), pp. 151-166, Interscience
Publishers, New York.
Riehl, M. L., Weiser, H. H. and Rheins, B. T. (1952) Effect of lime-treated water upon survival of
bacteria, J. Amer. Water Works Assn., 44:466-470.
-------
328
Seiberling, D. A. and Harper, W. J. (1955) HTST pasteurization for the control of psychrophilic
organisms in plant water supplies, J. Dairy Sci., 38, Abt.-M-47. 588.
Shuval, H. I., Cymbalista, S., Wachs, A., Zohar, Y. and Goldblum, N. (1966) The inactivation of
enteroviruses in sewage by chlorination, Proc. Third Internat. Conf. on Water Poll. Research,
Water Poll.Cont. Fed.
Smith, D. K. (1967) Disinfection and sterilization of polluted water with ozone, Report AM-6704,
Ontario Research Foundation, Toronto, Canada.
Solo, F. W., Mueller, H. F. and Larsen, T. E. (1970) Disinfection of sewage effluents, Progress
report to FWQA, U. S. Dept. of the Interior, Cincinnati, Ohio.
Starback, H. S. (1967) Iodine vapor generator and method of using it to treat and disinfect fluid,
U.S. Patent No. 3,352,628.
Stumm, W. (1958) Ozone as a disinfectant for water and sewage, Boston Soc. Civil Eng., 45:68-79.
Sullivan, R. L, (1970) Personal communication, U. S. Public Health Service, Food and Drug
Administration, Cincinnati, Ohio.
U. S. Public Health Service (1959) Construction Plans for Water Pasteurizer, Mimeographed
Report, R. A. Taft Water Research Center, Cincinnati, Ohio.
Wattie, E. and Chambers, C. W. (1943) Relative resistance of coliform organisms and certain
enteric pathogens to excess-lime treatment, J. Amer. Water Works Assn., 35:709-720.
Weidenkopf, S. J. (1958) Inactivation of type 1 poliomyelitis virus with chlorine, Virology,
5:56-67.
-------
CRITICAL REVIEW OF PAPERS ON TREATMENT PROCESSES
Karl Wuhrmann
Reviewing eight contributions of excellent and distinguished speakers is a delicate task considering
the fact that the most pertinent paper has not been presented at this symposium. I mean the basic
text, setting the specific goals and defining the special needs for water pollution control measures
in countries with arctic climates. Like in any other geographic region waste treatment has to meet
the requirements set by the quality standards which are or have to be imposed on the receiving
water bodies by legislation or on the grounds of ecological exigencies. As anywhere, the quality
standards of treatment plant effluents must be tailored to the local conditions in rivers and lakes,
and the technology of treatment can only be decided upon when these conditions are known.
Since this basic concept was lacking, the papers under review necessarily pointed in various
directions, according to the personal association of each author to the keyword of "cold climate."
Concerning the first fundamental question, namely "what has to be done?" three objectives have
been considered, i.e., the conventional problem of removal of organics, the abatement of
eutrophicating effects of wastes and the health hazards involved with sewage. No priorities have
been set. AH of these objectives might be important, however, at one place or another, and it was
justified, therefore, to cover the respective technology to a certain extent.
As to the second fundamental question "how can it be done?" the authors have mostly elaborated
on the problem of low temperature effects on process efficiency and design. Process efficiency is
undoubtedly an important item in the present context. I should say, however, that it has much less
weight in practical water pollution control than such banal items as sludge disposal or purely
mechanical problems of operating machinery in an arctic winter. As a microbiologist I am not
competent to deal with such technical matters. Having some experience, however, with small
treatment plants in the Alps up to altitudes of more than 9,000 ft., I wish to indicate that these
technical problems are by far dominating every other question which might arise due to low
temperature!
In the group of papers reviewed by Dr. Krenkel it has clearly been shown that the above
mentioned objectives one and three, i.e. removal of organics and reduction of health hazards, are
most imperative in the majority of Alaskan situations. It was made clear also that, besides a very
few large agglomerations where conventional planning of plants may be adequate, mostly small
population centers exist, requiring the utmost technical and operational simplicity for any
treatment installation.
I take the liberty to comment on the papers under the two headings:
• Basic considerations as to biological processes at low temperatures, and
- Practical observations and experiences with treatment processes.
329
-------
330
It is justified to begin with the papers presented by Eckenfelder and Englande and by Benedek and
Farkas, because the temperature effects on reactions and reaction rates in treatment plants (be
they of biological or chemical nature) are fundamental in our context.
It must be emphasized that we have to differentiate between:
- Temperature effects on the performance of the entire treatment plants or individual treatment
steps, and
- Temperature effects on single chemical or physico-chemical reactions or reaction chains,
including those occurring in organisms.
It would be a fundamental mistake to confound these two groups of temperature effects. In the
case of, for instance, a biological treatment plant, the temperature acts preferably on the
population dynamics of the entire system. The resultant effect on plant performance is, therefore,
not directly related to temperature functions (as for instance the Arrhenius equation) of
biochemical reaction rates. It is true, of course, that the population shifts in mixed continuous
fermentation systems are caused by temperature effects on growth rates and hence, on
biochemical reaction rates. The phenomenologically dominant result, however, is a change in the
competition situation of the individual species, leading to a so-called sociological adaptation of the
biocenosis (Wuhrmann, 1964). This change within the organism community must not necessarily
be associated with a shift in the overall fermentation performance.
In the paper of Eckenfelder and Englande temperature coefficients of numerous activated sludge
laboratory and pilot plant experiments, as well as of lagoons are compiled. The authors, and more
extensively Benedek and Farkas, stress the direct relationship of these observations with the van't
Hoff-Arrhenius equation for activation energy. It has to be recognized, however, that these
temperature factors are plainly empirical and should not be misused to anticipate the kinetics of
specific biochemical reactions. The temperature coefficient of endogenous and substrate
respiration of activated sludge has always been a favorite subject for studies and the essential body
of Benedek and Farkas' paper has many predecessors (Sawyer, 1939; Wuhrmann 1955, et al.). 0,0
values around 2 have been found consistently, from which E values of about 11,200 - 12,000
kcal/mol (temp, range around 20° C) may be calculated. Tnis figure is of little significance,
however, since we ignore completely the reaction that might be rate limiting within the observable
end result of oxygen consumption. It is noticeable that other entirely different reaction chains
such as the removal of a substrate from the medium (Benedek and Farkas) or the kill of bacteria
by disinfectants (Chambers paper) are also subject to temperature effects with values of 0,0 in the
range of about 1.6 to 2.5 (calculated activation energies around 8,000 - 15,000 kcal/mol). This is a
general indication that enzymatically catalized reactions might be involved. It is obvious, however,
that just because of this parallelism farther reaching conclusions as to the behavior of complicated
systems such as living cells or entire organism associations are not justified.
There is no objection against the determination of temperature coefficients of plant performance.
Such values are very useful in practice because they give at least a rough indication as to the
magnitude of safety factors to be considered in the dimensioning of biological reactors. It is
pertinent, however, to clearly separate 1) reactions in a biological treatment system which are
invariably affected by temperature changes according to thermodynamic principles, and 2) the
system's behavior as an entity under steady state conditions of practical operation. The classical
-------
331
example for the first group of reactions is sludge respiration with a temperature factor 010 around
2. irrespective of any other environmental conditions or population shifts. It is mandatory,
therefore, to consider this temperature factor in the dimensioning of aeration systems. Growth
rates as well as substrate resorption rates are subject to similar temperature factors. On a long term
basis of continuous operation and at slow temperature shifts, however, these temperature effects
will be largely hidden by a sociological adaptation of the sludge which may readily compensate for
the change in metabolic rates of the originally dominating species. Most sewage plants
demonstrate, therefore, amazingly small shifts of efficiency from summer to winter temperatures.
These small differences even disappear more or less completely at low sludge loads (0.1 - 0.2 kg
BOD/kg dry solids/day) because substrate supply and not temperature is then acting as the rate
limiting factor for sludge metabolism. I think there is no need for Eckenfelder's and Englande's
hypothesis, assuming sludge flocculation is an essential parameter for the magnitude of
temperature effects on the overall performance of conventional activated sludge systems. This
hypothesis is even contradicted by earlier experimental evidence such as Sawyer's observations
(1955) and the reviewer's own results (Wuhrmann, 1964). The idea of temperature-independent
rate limitations by too small a substrate supply is further confirmed by the skim milk experiments
in the paper of Koyama et al. and by the laboratory experiments described by Clark et al.
The essential practical conclusions are therefore:
1. Compensation for adverse effects of low temperature in activated sludge systems is readily
possible by low sludge loads. With domestic sewage the limit is in the order of magnitude of
0.1 to 0.2 kg BOD/kg MLSS/day. As was shown by the reviewer (Wuhrmann, 1964) and has
been confirmed with the skillful experiments of Clark et al., dimensioning has to consider the
lowest temperatures occurring.
2. Aeration rates have to be evaluated on the basis of the highest temperatures to be expected in
the system. Exigencies for low temperature will then automatically be satisfied.
3. Although lagooning might be a relatively cheaper investment than other systems, it is
doubtful it will serve the needs, emphasizing the fact that the highest treatment efficiency is
required in wintertime, i.e. at the lowest temperatures (see introduction of Gordon's paper).
The above conclusions lead to some technical consequences regarding sewage and plant
construction. From the operational point of view one of the main problems - especially with small
plants - is ice formation. Heat conservation is imperative, therefore, and requests a very compact
and condensed plant layout. Heavily exposed plants in the Alps of Switzerland are enclosed for
protection against snow and wind and for easier maintenance. The largest heat loss occurs in the
aeration basin due to evaporation and intensive exposure of the water to the atmosphere. As has
already been mentioned by Pick and others, surface aerators should, therefore, not be used. Very
much can be done in favor of safe plant operation with an adequate sewage system as has neatly
been shown in the paper of Koyama et al. He demonstrated how thawing may interfere with
treatment efficiency by increasing the hydraulic plant load and simultaneously decreasing the
sewage concentration and temperature. All three factors may work together and produce an acute
danger of sludge washout. The answer to this problem evidently is separate sewering, a concept
which is already adopted in the United States but still meets resistance in other places.
-------
332
Not many words have been devoted at this conference to the delicate topic of sludge treatment
and disposal, the paper of Balmer excepted. The problem is highly critical with small installations
where no machinery for sludge processing can be afforded. It is a favorable coincidence, however,
that small plants should be designed as very low loaded complete oxidation units, for reasons
already discussed. This concept automatically involves a minimum of excess sludge production and
a sludge quality which is inoffensive, creates no odor problems and dries rapidly (especially after
freezing). Under these conditions the conventional drying bed is probably the optimum solution
for sludge disposal.
The question of eutrophication by plant effluents has been discussed in the papers of Rosendahl
and Balmer in regard to phosphorus removal from sewage. This is of course an immediate problem
all over the world. Irrespective of the obvious question of whether nutrient removal from wastes is
of top priority in the Alaskan situation, the process described by Balmer raises an interesting point
worth mentioning: chemical processes such as precipitation or adsorption etc. have much lower
temperature coefficients (diffusion being mostly the rate-limiting process) than enzymatically
catalized reactions. Dimensions and operations of chemical purification plants are much less
affected, therefore, by temperature shifts than are biological units. This represents a considerable
advantage. In view of the high degree of purification required by the winter conditions in arctic
rivers, it is questionable, however, whether exclusively chemical processes as described by Balmer,
are sufficiently effective in regard to removal of dissolved organic compounds. It is also common
experience that handling of excess sludge from precipitation units can be a difficult task. I am of
the opinion, therefore, that the application of chemical precipitation or flocculation processes
under critical operation conditions (small plants, low temperature, unskilled personnel, etc.)
should be considered with caution.
REFERENCES
Sawyer, C. N. (1939) Factors involved in prolonging the initial high rate of oxygen utilization by
activated sludge - sewage mixtures. Sew. Wks. J., 11, 595.
Sawyer, C. N., Frame, J. D. and Wold, J. P. (1955) Revised concepts on biological treatment Sew
Ind. Wastes, 27, 929.
Wuhrmann, K. (1956) Factors affecting efficiency and solids production in the activated sludge
process, Biol. Treatment of Sewage and Industrial Wastes, Vol. 1, 49-65, Reinhold Publ.
Comp. New York.
Wuhrmann, K. (1964) Bibl. Microbiol. Fasc., 4, 52-64.
Wuhrmann, K. (1964-68) Hauptwirkungen und Wechselwirkungen einiger Betriebsparameter im
Belebtschlammsystem. Ergebnisse mehrjahriger Grossversuche, Schweiz. Z. Hydro!., 26,
218-270, see also Adv. in Water Quality Improvement, Univ. Texas Press, p. 143.
* U. S. GOVERNMENT PRINTING OFFICE; 1912 O - 454-699
------- |