A MATHEMATICAL MODEL OF PRIMARY PRODUCTIVITY AND by Lorne Gordon Everett

A MATHEMATICAL MODEL OF PRIMARY PRODUCTIVITY  AND by Lorne Gordon Everett
A MATHEMATICAL MODEL OF PRIMARY PRODUCTIVITY AND
LIMNOLOGICAL PATTERNS IN LAKE MEAD
by
Lorne Gordon Everett
A Dissertation Submitted to the Faculty of the
DEPARTMENT OF HYDROLOGY AND WATER RESOURCES
In Partial Fulfillment of the Requirements
For the Degree of
DOCTOR OF PHILOSOPHY
WITH A MAJOR IN HYDROLOGY
In the Graduate College
THE UNIVERSITY OF ARIZONA
197 2
THE UNIVERSITY OF ARIZONA
GRADUATE COLLEGE
I hereby recommend that this dissertation prepared under my
direction by Lorne
entitled
Gordon Everett
A MATHEMATICAL MODEL OF
PRIMARY PRODUCTIVITY AND
LIMNOLOGICAL PATTERNS IN LAKE MEAD
be accepted as fulfilling the dissertation requirement of the
degree of
Doctor of Philosophy
D ssertation lirec or
K/r/ovL-Date
After inspection of the final copy of the dissertation, the
following members of the Final Examination Committee concur in
its approval and recommend its acceptance:*
Yfr/9
-27
,57/ 3//7
?
•
This approval and acceptance is contingent on the candidate's
adequate performance and defense of this dissertation at the
final oral examination. The inclusion of this sheet bound into
the library copy of the dissertation is evidence of satisfactory
performance at the final examination.
STATEMENT BY AUTHOR
This dissertation has been submitted in partial
fulfillment of requirements for an advanced degree at The
University of Arizona and is deposited in the University
Library to be made available to borrowers under rules of
the Library.
Brief quotations from this dissertation are
allowable without special permission, provided that
accurate acknowledgment of source is made. Requests for
permission for extended quotation from or reproduction of
this manuscript in whole or in part may be granted by the
head of the major department or the Dean of the Graduate
College when in his judgment the proposed use of the
material is in the interests of scholarship. In all other
instances, however, permission must be obtained from the
author.
SIGNED:
ACKNOWLEDGMENTS
The research described herein was made possible by
a grant from the U. S. Department of the Interior, Bureau
of Reclamation, Region III. The Computer Center at The
University of Arizona and The University of California at
Davis provided the computer time in the analysis.
Special gratitude is expressed to the research
team who volunteered their services to make this comprehensive study possible: K. S. Hanks, R. D. Staker, G. C.
Slawson, Jr., and C. L. Constant III.
The author wishes to thank Dr. Charles Stockton
for his knowledgeable help in the computer analysis.
iii
TABLE OF CONTENTS
Page
LIST OF ILLUSTRATIONS
vi
LIST OF TABLES ix
ABSTRACT CHAPTER
1. INTRODUCTION 1
Objectives Related Investigations 2.
8
8
THE AQUATIC SYSTEM
Conceptual Model
Review of Functional Relationships
in Modeling
The Nutrient System
The Phytoplankton System
The Zooplankton System
3.
44
Locations and Water Quality
Considerations Recreational Uses 44
51
4. EXPERIMENTAL DESIGN AND METHODOLOGY
iv
53
53
59
59
64
State Estimation Sampling Dates Sampling and Measurements Primary Productivity--C Method
Data Description Temperature Dissolved Oxygen (DO) Soluble Salts Electrical Conductivity (EC)
15
17
23
34
40
THE STUDY AREA
5 . RESULTS AND DISCUSSION
15
67
67
67
69
71
73
TABLE OF CONTENTS--Continued
Page
6.
Hydrogen Ion Concentration (pH) Chloride, Magnesium, Potassium,
Sodium, and Calcium
Phosphate
Nitrate
Bicarbonate
Iron, Manganese, Zinc
Copper
Sulphate
Light Intensity
Zooplankton
Other Zooplankton Studies on Lake Mead . The Rotifera
The Crustacea
Cladocera
Copepoda
Ceratium
Asterionella
Primary Productivity (PPR)
96
96
98
103
103
109
SYSTEM ANALYSIS
Multivariate Analysis
Stepwise Regression Analysis
Principal Component Analysis
Testing the Model
7.
CONCLUSIONS
73
76
76
76
81
81
81
85
85
87
89
90
93
109
115
120
127
130
130
139
141
Discussion
Future State Set Analysis
APPENDIX A. ZOOPLANKTON DISTRIBUTION
APPENDIX B. STEPWISE REGRESSION PRINTOUT
REFERENCES
144
146
LIST OF ILLUSTRATIONS
Page
Figure
4
1.
Location of Salt Springs
2.
Simplified Aquatic Model
16
3.
The Aquatic Ecosystem
18
4.
Ecosystem Flow Diagram
19
5.
The Biogeochemical Cycle of Nitrogen
Compounds in Aquatic and Other
Environments 25
6.
The Biogeochemical Cycle of Phosphorus . . 26
7.
Simple Motile Green Alga, Chlamydomonas • • •
33
8.
Phytoplankton Growth Rate as a Function
of Incident Solar Radiation Intensity
and Temperature 35
Depth to Which Light of Various WaveLengths Penetrates, Expressed as
Percentage of the Surface Light 36
Absorption Spectra of Three Types of
Chloroplast Pigments: (a) Chlorophylls;
(b) Carotenoids; (c) Phycoerythrins and
Phycocyanins 38
9.
10.
11.
Flow Diagram for a Simulation Model of
Rotifer Population
12.
Map Showing the Colorado River Basin from
Lake Powell to the Mexican Border
13.
Salinity Releases at Hoover Dam
14.
Map of Lake Mead Showing Basins
15.
Salinity Values at Grand Canyon
16.
River
Salinity Values at the Virgin
vi
43
45
46
47
49
50
vii
LIST OF ILLUSTRATIONS--Continued
Figure
Page
17.
Four Classes of State Sets Investigated
18.
Independent Variables Investigated
19.
Location Map of the Sampling Stations on
Lake Mead
20.
Temperature Cycle at Beacon Island
21.
September and January Dissolved Oxygen
and Temperature Profiles
22.
23.
54
September and January Soluble Salt
Concentrations
56
61
68
70
72
January and September Values of Electrical
Conductivity
74
24.
September and January pH Changes
75
25.
September and January Chloride
Concentrations
77
September and November Calcium Concentrations
78
September and April Phosphate Levels
(Po )
4
79
28.
September and April Nitrate Concentrations . .
80
29.
September and January Bicarbonate
Values 82
30.
February and June Zinc Concentrations • • • •
83
31.
February and June Concentrations of
Copper 84
26.
27.
32.
September and January Sulphate
Concentrations
86
33.
Phytoplankton-Zooplankton Growth Patterns • •
88
34.
Seasonal Zooplankton in Lake Mead
Compared to Pennak (1946)
92
•
viii
LIST OF ILLUSTRATIONS--Continued
Figure
35.
36.
Page
Total Number of Keratella with January
(r----1) and June (11.701) Values Shown 94
Total Keratella Counted at Las Vegas
Wash and Temple Bar
37.
Seasonal Peaks of Daphnia and Bosmina • • •
38.
Total Numbers of Calanoids and
Cyclopoids
95
97
99
39.
January Values for PPR and Cyclopoids • • • •
100
40.
Number of Cyclopoids Recorded at Las
Vegas Wash and Temple Bar
101
Total Number of Ceratium with January
(F 7) and June (cimm) Values Given
102
41.
-
42.
Space-Time Distribution of Primary
Productivity
106
43.
Dominant Eigenvector of the 31 Variables . . 126
44.
Minor Eigenvector of the 31 Variables . . . 128
45.
Temporal Changes in Algal Growth Rates . . . 132
46.
Spatial Distribution of PPR and
Phosphate (September, 1970)
135
LIST OF TABLES
Table
1.
2.
Page
Average Annual Total Dissolved Solids
Concentrations
at Selected Stations
(1960-1970)
2
Classification of Lakes by
Eutrophication 6
3.
Comparison of the Two Schools of Thought
4.
Grazing Rates of Zooplankton
39
5.
Experimental Design
60
6.
Survey Dates and Seasons
7.
Primary Productivity at Beacon Island
8.
Autotrophy (Phytoplankton)
9.
Parameter Index Table
.
.
29
62
•
•
•
•
104
107
116
10.
Principal Component Analysis
121
11.
Correlation Matrix 122
12.
Per Cent of Variance Explained by Each
Eigenvalue 125
ix
ABSTRACT
The temporal and spatial changes in chemical and
biological properties of Lake Mead have been investigated,
thereby indicating the sources of water pollution and the
time of highest pollution potential. Planktonic organisms
have been shown to indicate the presence of water problems.
Macro- and micro-nutrient analyses have shown that primary
productivity is not inhibited by limiting concentrations.
A mathematical model has been developed, tested with one
set of independent data, and shown worthy of management
utility. Although the model works very well for the Lake
Mead area, the physical reality of the MLR equation should
be tested on independent data.
CHAPTER 1
INTRODUCTION
Progressive increases in concentration of dissolved
solids in the Colorado River from Lake Powell to Imperial
Dam seem to alter plankton dynamics and biological productivity of the river. Also, changes in biological productivity and nutrient concentrations occur within the same
reservoir. Quantification of the relationship between
physical and biological components of the system are necessary to diagnose eutrophication trends and the carrying
capacity of the reservoirs and river reaches.
The Colorado River in the Lower-Basin States is
highly saline compared with other major rivers in the
world. Between the Green River in Wyoming and the Imperial
Dam, there is a 2-3 fold increase in dissolved solids
concentration in the River (Table 1). High salinity of the
water and the lack of a reliable model to predict future
trends are major constraints in deriving alternative
developmental plans for urbanization, recreation, and water
uses in the Lower-Basin. The Department of Interior estimated that domestic, industrial, and agricultural activities
in the Upper-Basin States has caused a rise in salinity of
about 240 ppm at Lee Ferry (Iorns, Hembree, and Oakland,
1
2
-
fh-
H
-0
I
4)
tih
-I-)
H
›.-n
H
..,
;-1 P
a) ;.., 05
41 a) -0
i
iI
-4-)
i
I
1
I
i
P 0
-
HNONN-00-1-.
en
cn
en
N
0
ch
n.0
c.",
(n
cn
cn
en
N
N
.,
-
.-
n1`
.-,
dr`
H H H H H H H
0 H 0
Z
PZ1
Q 4.)
a) Z
HQ) ;.1
a> 5 c)
t/)Q)i
0.) ..,
-P-0
a5 0 a)
(,c5
0
u) W ^
P
(5
O Q) 't
.H -p 05
-1-) (5
cc5 A-) a)
• u) Z
-P 1
0 H •'
0 (CI In
O ;-I ('5
0 0 ID°
O '0 0
H
(z
H
H
(cs
0
E I
cl) 05 N
P., (P) •,--1
N-
r`--
N-
E
H
0
c0
H
.n.0
N
n.0
0
ENNHohNHoh-d-1-d-1NN
(X)
c0
N-
00
ch
c0
c0
00
oh
(3"h
.4
H
;-1
«5
a)
0
,-- E I
O5 N
Ct5 (: .r.-1
ai
<
H
(-,1
n.0
c>
(7\
n.0
C›
..0..0
H
CO
cn
trl
N-
.C,
n.0
IN--
1-fl
tr\
N
CT
\-0
---i'
N-
Ch
CO
VD
N.
-1-1
CO
IN-
cvl
--t
IN-
Ill
I'-.
nD
Cr\
Cr\
n.0
\-0
N-
N-
\ -0
NN-
n.0
s:)
iih
H
c0
.0
H
N-
Ch up
.c:,
c0
H
th-
N-
H
Ill
H
Cg
%...0
N--
- \ .0
N
CD
‘..0
H
Ce\
k..0
N
oo
cn
th-
oo
N
cn
..o
if\
H
ch
co
Cvl
Cel
C. n
c0
c.) a>
ÇLI
Cf)
th
t $ •P CS
(1)
(1)
Z
1-1 05 ,...E
H
N.
Ilh
N-
CI
0,
i
H
0 OC N
IN-
0 \ 00
IN-
Cg
0
O "0 ^
0 Ci .1-1
'.0
n.0
n.O
IN-
CO
CO 0 ;-i
-1-) 0
.4
rCs 0
4) 01) .H
CO %
HQ)0
O
0
't ON
N
In aq '0
-...t,.1D
0,
0
cr.c)
g >1 .r1
N
u)
00
Cr\
Cr\
(5
H
(5 P
en
c::,
•,.-1 I
..,
uh
cy.
n,0
n.0
th-
(5 .4
.,
(0) 1 0(50
H
H
H ---, 0
ai 00
-3-) N,c)
O CN 0
cn
0 ..i N
uh -.1-1H
cl
a)
-H
0\
H
ce)
CI
H
EH H 4
I. ' --
I -I-, COIAI,-
Ir \
ON
CO
Ir \
HO
RI .(D g
O 0\ 0
g
Z ...... 0
g
0 0 0
IN-
0\
.CD
CI
V.')
CI
<
(-)
c\I
c)-\
U) g
c)
N--
H
-- 0
ce-\
n
coq
n
cq
en
Q) z a)
-H
bi) 0 ;..1
cs • H a) •
-• 1-.) 4-1 cl
(l6 Z r•-
(.5
Q)
-4-) 0 (:;r
<4 u) c.) H
E
0
0) 0
H
( 0
c7n
(11
-t-i
n.,0
-I-.
c,
,...1
0\
CO
CI
0-1 CO
en
L/1
CO
Ch
Ch , •.0
Ch CO
Ch n..D
CO
n
cn 00
H
n0e
--1-1anIA
--ti-1-1-- cn
00
H
Q)
H
;-1
rû
CC5
Z
El
CD
N
c...\
-st
mp
ON
N-
00
CD
I'D
QD
C:)
n.0
N.
ch Cn
ON
0\
ON C \
C7\G" CN
H H H H H H H H H H H
\ .0
.SD
Ch G\
IA
3
1965). Since that time, salinity in the lower Colorado
has increased about 300 ppm. Part of this increase may be
attributed to changes in sediment dynamics, growth of
phreatophytic-type vegetation, and riparian growth. There
are large areas along some reaches of the river where
deposited sediments have resulted in major ecological
changes and to some extent provided excellent conditions
for vigorous plant growth. Salinity increase in the river
is attributed to two mechanisms: salt loading, and salt
concentration. Source points of loading are as shown in
Figure 1. Increased salt load is also due to urban, industrial, and agricultural developments and the discharge of
pollutants associated with these activities. Salt concentration results from evaporation, transpiration, and
diversions.
Changes in salinity and direct and indirect introduction of domestic and industrial effluent into the system
relate directly to biological productivity and eutrophication trends. It is anticipated that future developments that add new types of inputs through use and/or
result in diversion have a direct relationship to pollution
trends in the system, its social and physical carrying
capacity, and consequently, its regional and national
importance as an economic, social, and political unit.
The man-made reservoirs and river reaches represent aquatic
systems with extreme conditions. States of water bodies
4
NUMBERS
=
TONS PER
YEAR
Figure 1. Location of Salt Springs
5
vary from oligotrphic to eutrophic in the same lake at
different locations and different times. Lakes and other
surface waters are frequently divided into one of two
types, oligotrophic or eutrophic. It is generally agreed
that oligotrophic lakes are relatively unproductive and
receive small amounts of aquatic plant nutrients, while
eutrophic lakes are highly productive and experience high
fluxes of aquatic plant nutrients (Table 2). The word
eutrophication has many problems in its definition because
of recent usage. Originally the term described the general
nutrient condition in German bogs. The term later was used
to describe a stage in the life span of a lake. Today the
term is used to describe the flux of aquatic plant nutrients
and/or the amount of plant or animal production. Still
others use the term to relate nutrient flux to water quality. In general, we can say that eutrophication is an
aspect of the natural process of the aging of an aquatic
system that can be accelerated due to man's developmental
activities.
Lake Mead is the largest surface water body in the
basin and it represents a system with wide variability in
limnological conditions. The present states of the Lake
need analysis to depict the influences of past activities
and to develop a sound base for forecasting the effects of
developments.
6
U)
a)u) u)
g-,
0Vg
H
0
u) g
o
4
a) .,--1
H 0
-H
-T-1 (.)
a)
P.,
a)
-0
C.)
H
-H
0
4
mS
bo
bo
H
HO4
o -H
o
0
O a) 5
H E -H
-0 4 H 0 H
bi)
bl)
cti H -H
H
-H
4 OP o
(I)
11)
i-
0 0
"0 -H
u)
0
a)
g
H
cp
OH0
boo0
P.,
H
4
H
+) 4
-H
u2COU)0 5
es
tr)
CI)
0
a) -0 4
H c:
H El
es
;-1
u)
b0
.H
(.5
hl)
.H
Z
g "C3
g
z
H ..1
O'
g
c)
H g.,
-H O Z
.,-1'CU)H 0
g
0
.H 4-1
0
CF-..
0
4) 41-1
— OE
tn
0
0 0 'CI
-H C>6
oOU)
g., H
cv
0
Q
o
0 0
gi •r-1
• -p
c) 0 u)
u) -H0< "CS
U) "E
E
;--, o
0 +> 0
H
Cll H H H -H H 0
1-4 H
'CS
-1-4) $..1
4 -P o 4
0 •H
pp.
bi)
,
(..)OU) ,.0
H
.1-1 4-nHO0 0 (..)
Z
i4 4
c24 i--.
aq 0-1 ---'
0
-4-)
a)
u)
+>
+
+
4
-P E
H
0
o
0
o g
E
0 S-1
a)
0
a)
—
u)
E
O 0
HOH
O HQ)
tr/
Z
E
+>
^ -H
-p
COU)
0 a)
es
-0 .H
-H
rn HO
a) -H
H 0
a)
ti)
-HO-) .
0
-H
0
-4-)
a)
o
0
bp E
0 o
o +)
-P
4 o
-4-) 4
;-1
-H R4
0
0 U)0
U)
A
0-,
a) p
o-H
0)0g
-P -H
U) 0
E
-P
0
4
-pr--I(1)
-H 0
H u)
ça+
. .0 f=-1
CO E-I
R-I 4
O
o
44
1.-
-P
---,
H
o
a) 4-1
-P
p-,
0-1
0
Z
a)
Os-404
0 0
g H
-H "CS
g -P
-H -H
4
-P
4 0
0
.H
o
o
zo
1-4
i-1
H
ti)
o
c\I HObi)OU)
o 00
OH-H 0
0.,
CI c.)
p.,
f:
H
0
ct
Z
0
0
0
O
U)a5
S--1
(cS
P,
u)
as-p -p
o
o
bl)
H
oE
LI
-4-)
0
-0
0
0CtS
o)
4-1
H
-0 cci
g u)O
a5
g
ial
0 0
4-1 5
0
u) u)
U)
g
a)
e,
g
-H
H
g
u)
"C4 0
0
+)
• -H
0
;-I
4
C6• —I
a)
E
;-4
+>
-4-)P. -P
-H
ra.,
0
H
III
g
H
o —
u)
g
-P
•
a)
4- 0
H P.4
;-1
bf)
.H
lal 0
•
O
g
"0
O
4
Z
bi)
;-4
0
4-1
0
-0
P-
tn o
HOH
E
Z 0O
0
'OU)
Cf)
P,
u)
Pa
• H
C=)
;-i u)
(CS .,-.1O
"0
0
-H
ri
a)
u)
ZP
4
c\1
0
H
0
ti)
-P
g
o
-P
u)
1es
H
rH
Q
P-1
g
a)
'ID
el
o
a)
V
7
cz,
8
Objectives
A generalized objective of this study is to analyze
the biotic and abiotic components of Lake Mead, and to
develop methodology for diagnosing changes in biological
productivity as it relates to physical and chemical changes
in the aquatic system.
Specifically, the objectives were:
1.
To establish a water sampling procedure for assessing temporal and spatial changes in chemical and
biological properties of the river-reservoir
system.
2.
To measure primary productivity rates (ppr) and
conduct counts of plankton populations at several
sampling locations in Lake Mead.
3.
To determine macro- and micro-nutrient concentrations at each sampling location.
4.
To derive functions relating ppr to physical and
chemical changes in the system.
Related Investigations
Since the completion of Hoover Dam in 1936, a series
of extensive investigations have taken place in Lake Mead.
Starting in 1937, the National Research Council (NRC)
studied the complexities of the density currents of the
Lake and published three volumes (NRC, 1949). The findings
continue to be used as reference on sediment dynamics in
9
the Lake. However, since the filling of the reservoir,
circulation and sediment transport patterns have changed.
Smith, Vetter, and Cummings (1948) conducted a comprehensive survey of the sedimentation process in Lake Mead.
Their study dealt with sediment distribution rates,
texture, bacteriology, and chemistry. Anderson (1950) and
Anderson and Pritchard (1951) began a massive effort to
define the physical limnology of Lake Mead. Theyconcluded
that salinity and temperature over one month must be
treated as conservative and non-conservative properties
respectively in circulation studies. The Virgin Basin was
shown to act as a large "mixing bowl" that dampens seasonal
variations in salinity of the Colorado River waters as they
flow into Lake Mead. Below Virgin Basin the water was
reported to be nearly uniform with respect to salinity.
The increase in flow from Las Vegas Wash has recently been
shown by Everett (1971) and Slawson (1972), using crossspectral analysis, to alter the salinity values at Hoover
Dam. The seasonal and trend increases in salinity at Las
Vegas Wash have been shown to immediately increase the
salinity values at Hoover Dam. The lag time between the
increases at Las Vegas Wash and Hoover Dam was less than
one month. The U. S. Weather Bureau (1953) supported the
circulation and sedimentation studies by establishing the
wind patterns over the reservoir.
10
By the end of the 1950's, the emphasis in Lake Mead
had shifted from sedimentation studies to water-loss investigations. Harbeck (1958) expanded the work of Anderson
and Pritchard (1951) and estimated evaporation losses at
seven feet of water per year.
The research interest in the 1960's was directed
toward the water pollution problems developing in the
Boulder Basin Region (Figure 14, p. 48). The Bureau of
Reclamation (1965) investigated the water quality of Boulder
Basin during the months of April and May. Based upon dissolved oxygen (DO)
'
CO
2'
pH, electrical conductivity (EC),
and temperature, they concluded that the impoundment
created by Hoover Dam did not adversely affect the water
quality of Lake Mead. The conclusion is a poor generalization since the effect of Hoover Dam is aggravated by the
Las Vegas Wash inputs. The Environmental Protection Agency
(EPA) (1967) gave a general indication of the problem
developing in the Las Vegas Wash area. However, their
findings were based upon phytoplankton counts. Numbers of
individuals are a poor parameter to quantify water pollution since it has no relation to the primary productivity
rate. In 1968(a), EPA conducted a general biological
survey of the Lower Colorado River. Using the number of
individuals, i.e., algae/ml they indicated that Las Vegas
Wash had the highest number of algae/ml and that the number
of algae/ml decreased with distance from Las Vegas Wash.
11
This report has since been shown to be incorrect in its
interpretation. Reports by Everett and Qashu (1971) using
sensitive radio-active 14 C primary productivity rates (ppr)
have shown that photosynthesis increases toward Hoover Dam
and that the problem is not located solely in Las Vegas
Wash.
In 1970, the Bureau of Reclamation published another
review of the water quality in Lake Mead. Using chlorophyll
as an indicator, they showed that the chlorophyll concentrations were much higher in Las Vegas Wash than Boulder
Basin. The highest values were obtained in May. The problem with chlorophyll is that it works reasonably well for
relatively pure cultures of one type of organism. However,
with mixed populations of organisms, particularly where
there is a shift in frequency of species, it is found that
frequently the correlation between the chlorophyll content
and the total numbers or mass of algae is very poor (Lee,
1970). The Bureau (1970) data show the problem to be
reduced by August when in fact it should be intensified
with the warming of the surface water (Everett and Qashu,
1971).
The Lower Colorado Region Comprehensive Framework
Study (1971) published eighteen volumes covering interests
and problems in the Lower Colorado River system. Volume
XV, dealing with water quality, pollution control, and
health factors, used number of algae/ml as their rationale.
12
They wrote that although Las Vegas Wash measured 7.0 mg/1
of nitrates--140 times greater than the 0.05 mg/1 criterion
for streams entering reservoirs set by the National Committee on Water Quality Criteria (EPA, 1968b)--"conditions
within the main body of the Lake are acceptable."
In December of 1971(a), the EPA seemingly under
considerable pressure, analyzed the problem in Las Vegas
Wash using chlorophyll "a" as the criterion for water
quality. They showed that the pollution decreased away
from the Wash. Their algal growth potential, which they
did not describe or give values to, showed that waters
below Hoover Dam had a higher growth potential than Las
Vegas Bay. They then said that this growth rate demonstrated that low phosphate concentrations limit algal
growth in Lake Mead. This is obviously contradictory since
higher phosphate values are found in the Wash area. The
EPA conclusions are based upon chlorophyll "a" as an
indicator of algal growth rate. Chlorophyll "a" at best
is a measure of biomass and does not reflect growth rates.
EPA failed to recognize that Las Vegas Wash did not behave
independently of the rest of Boulder Basin.
Riesbol, Minkley, and Kilmartin (1971) completed
a cursory study in Lake Powell for the Bechtel Corporation.
The study generally looked at plankton, fish, and the need
for hydrodynamic studies. Anderson (1971) published the
first report of a "Collaborative Research on Assessment of
13
Man's Activities on the Lake Powell Region," which is an
integrated chemical, biological, and hydrodynamic study
leading to a predictive model. Using 14 C techniques,
sampling in the first year was done at one, two, four, six,
and eight meters. A considerable loss of information in
defining the vertical ppr profile could be expected when
the depth of sampling is arbitrarily chosen.
Chemical, bacteriological, and biological investigations have been done by Everett et al. (1970-71) in the
Grand Canyon from Lee Ferry to Diamond Creek. Several
sources of chemical and bacteriological contamination were
found. The research is continuing in an effort to develop
models for diagnosing bacteriological and chemical changes
due to the intensity of river use for recreation.
The Environmental Protective Agency (1971b, c, d,
e) published a summary and three appendices on the Mineral
Quality Problem in the Colorado River System. On the basis
of these documents at the Federal-State Enforcement Conference on the Colorado River, Las Vegas, Nevada, February
15-17, 1972, the EPA was unable to obtain an agreement with
other State and Federal Agencies on a salinity standard.
The states made it quite clear that sound basic research
on the functional relationship of salinity to water quality
problems was needed before any management decision could be
made. Our participation in that Conference was to show
that an understanding of the ionic species change,
lk
accounting for the salinity problem was the major
concern. The physiological damage to aquatic and riparian
organisms varies with the ionic species, and not as the
total salinity changes. The source of the ionic species
could be more easily identified by understanding the
composition of the salinity values. Our final comment at
that Conference was to show the inadequacy of basing
standards on a mean value. The extremes associated with a
mean value are primarily responsible for biological damage.
CHAPTER 2
THE AQUATIC SYSTEM
Conceptual Model
The impact of water resourcesdevelopment projects
on aquatic eco-systems, measured in terms of accelerated
eutrophication and resulting economic and social losses,
has become a matter of rising concern both in the United
States and abroad. If we consider the potential combined
effects of increased domestic sewage, salinity increases,
and agricultural drainage effects we are led to conclude
that the environment in the Lower Colorado River would
experience a greatly accelerated eutrophication. There is
little doubt that a serious need exists for methods and
tools which will permit assessment of the shifts in rates
of eutrophication and in the merits or demerits of alternatives for control of eutrophication. Realizing the need
for such a capability in dealing with eutrophication, we
find it necessary to understand biotic and abiotic relationships in the aquatic system. A simplified aquatic
model (Figure 2) deals with a nutrient source, primary
producers (phytoplankton), and primary consumers (zooplankton). Usually very general parameters such as temperature are the variables of interest. Most of the inputs
15
16
17
are lumped, i.e., fish grazing is represented by one
factor. Models of fish population dynamics have been
expanded by Patten (1969).
A more complete model of an aquatic ecosystem is
given in Figure 3. The higher plants have been omitted.
Although we can conceptually "understand" the paths in
Figure 3, obtaining functional relationships is difficult
because of the lack of complete simultaneous data acquisition.
The flow diagram (Figure 4) shows the interaction
(I) of the components. Limnologists are aware of the
complexities in trying to represent the various nutrient
phytoplankton and zooplankton systems as a black box. There
are many chemical and biological phenomena taking place in
the system that are not fully appreciated and are not presented (i.e., phytoplankton affinity for different nitrogen
compounds).
Review of Functional Relationships
in Modeling
Considerable work is reported on the development of
eutrophication models. Some of the early work was done by
Davidson and Clymer (1966) using analog and digital
approaches. These authors had to make many assumptions:
1. All physical, chemical, and biological variables
were uniformly distributed through the volume of
water of concern.
O
O
r2n
E—
19
o
cN
CT1
co
•••••n
(NJ
1,•n•
cD
CNJ
20
2.
All variables were averages of 24 hours since
diurnal frequencies were not of interest.
3.
All species present can be lumped into two classes:
phytoplankton and zooplankton, each having assigned
constants over all species, ages, and times of
year.
4.
There was one critical nutrient.
5.
Growth and reproduction need not be distinguished.
6. Illumination and water temperature have sinusoidal
annual cycles.
Davidson and Clymer then developed a set of equations for a
mathematical model. Their generalized equation is:
P = K
1
(N limit ) - K Z - K - K T
4
2
3
(2.1)
where
P = phytoplankton population density
Z = zooplankton population weight density
N = nutrient concentration (limiting element)
T = water temperature °C
K 1 = phytoplankton growth rate due to photosynthesis
K
2
= death rate of phytoplankton eaten by zooplankton
K 3 = all other contributors to death rate of phytoplankton
K 4 = rate of energy needed for respiration.
21
The result of their efforts showed considerable opportunity
and desirability for ecosimulation studies.
Cole (1967) at Battelle Northwest expanded upon the
work of Davidson and Clymer. Temperature, death rates, and
grazing rates were varied in the analysis. The Water
Resources Engineers, Inc. (Chen and Orlob, 1968) published
a report on a proposed ecologic model for a eutrophying
environment. They used the rate law of chemical kinetics,
"the rate at which a specific reaction proceeds is some
product function of its reactant concentrations." In the
case of photosynthesis (a), they expressed the rate as
follows:
da
= K x f(A) x f(I) x f(CO )
dt
2
f(N) x f(P) (2.2)
where K is a coefficient and f(A), f(I), f(CO 2 ), f(N), and
f(P) ... are respectively functions of advection, light,
carbon dioxide, nitrogen, phosphate, etc.
In particular cases, the concentration of some
factors may be relatively large so that the level of the
abundant material may be treated as practically constant.
This could result in an apparent rate equation of only a
single concentration factor. This factor becomes "limiting"
according to Liebig's classical "Law of the Minimum" and
Blackman's concept of "Limiting Factors." Odum (1963) has
extended Liebig's Law of the Minimum to include the
22
limiting effect of the maximum (i.e., an excess of chloride
could be toxic or inhibit algal growth).
DiToro, O'Connor, and Thomann (1970) were able to
develop a hybrid phytoplankton model at Manhattan College,
Bronx, N. Y. DiToro followed the equations presented by
Riley (1965) and Steele (1965). The principle of conservation of mass was the basis upon which the mathematical
development was structured. All equations were developed
in terms of adding to or subtracting from a source or sink.
Chen (1970) made a plea that a "multidisciplinary"
background be required to bring scattered bits and pieces
of accumulated information to bear on the broader problem
of management of aquatic environments. He reviewed the
Water Resources Engineers' approach and suggested some
equations for a eutrophication model. For example, he
suggested that biomass of phytoplankton is transported by
movement of water. In addition, phytoplankton increase is
attributed to reactions depending on light, temperature,
and nutrient conditions. Also, phytoplankton decrease as
a result of continuous respiration, settling, and grazing
by zooplankton. Terms to account for these effects were
presented in the differential equation:
d
(VPi)
dt
- TAEdpi+ (H -r-s) piv _ YzgpfizV
dx
where V is volume of the element (L 3 ); P i is mass concentration of algae in group i (M/)); t is time (T); T is
23
total advective mass transfer of biomass (M/T); A is crosssectional area of the element (L 2 ); E is effective diffusion coefficient (L 2 /T); x is distance (L); Hi is specific
growth rate of algae in group i (l/T); r is per cent of
mass respired per unit of time (l/T); s is per cent of mass
settling per unit of time (l/T); Y z is yield coefficient of
zooplankton (M/M); g is specific growth rate of zooplankton
(l/T); P. is preference factor for algae, group i (dimensionless); and z is zooplankton biomass concentration
(WO). Each element is a part of a discritized network
having both vertical and horizontal dimensions. In this
way stratification and horizontal spatial variabilities can
be included. For any given time interval, each element is
considered to behave as a completely mixed reactor, allowing the mass of water quality constituents to be transported in and out by both advection and diffusion.
The Nutrient System
The essential elements for phytoplankton development listed by Hutchinson (1967) include C, N, 0, P, S, K,
Mg, Ca, Si, Na, Fe, Mn, Zn, Cu, B, Mo, Co, V, and the
chemical species of each element (i.e., NO 2 ). Lists that
include certain vitamins such as thiamin, cyanocobalamin,
and biotin are documented. The relative importance of
these nutrients has not been fully demonstrated.
24
Contributors to the nutrient core are natural
sources and man-induced waste loads (Figure 4, I i ). Although industrial and agricultural effluent are included,
the main source of nutrient is the domestic sewage in
either the raw or treated stages. The irony of expensive
secondary treatment is that organically bound nutrients
are liberated by the bacteria through the process of decomposition. Both the nitrogen cycle (Figure 5) and the
phosphorus cycle (Figure 6) in aquatic systems have been
described by Russel-Hunter (1970).
Phosphorus, as a major contributor to eutrophication, is a prominent element in the hierarchy of water
quality problems. Widely quoted figures given by Sawyer
(1947) have contributed to popular conception. His study
in the Madison Survey supports the belief that phosphorus
is the key element in determining biological activity in a
body of water. He showed that lakes producing nuisance
blooms had average concentrations of organic phosphorus in
excess of 0.10 ppm (100 !lei). Inorganic phosphorus however, could create nuisance conditions in amounts in excess
of .01 ppm (10 tig/1). Sawyer arrived at his conclusions
concerning the key effects of phosphorus by a bioassay
comparison to nitrogen. He showed that extensive algal
growths were produced under laboratory conditions with
plentiful supplies of phosphorus and deficient supplies of
nitrogen. Undoubtedly nitrogen fixation, either bacterial
25
Gaseous nitrogen /
/
in atmosphere
/ N2
:=1
\\
---
—
Photochemical
\
Fixation
/N2
Nitrate\
\.,..sediment
.„.‘"<"°•.
volcanic activity
ammonium-nitrogen
N-F.= nitrogen-fixing bacteria
D-N = denitrifying bacteria
N-R = nitrate-reducing bacteria
organic nitrogen
inorganic nitrogenl
—
—
plant nutrients \
Figure 5. The Biogeochemical Cycle of Nitrogen Compounds
in Aquatic and Other Environments -- After
Russel-Hunter (1970).
26
loss to sediments
Dissolved
.)pOrthophosphate Ions
4
:
Fertilizer '
Use
--
Fossil Bone
Deposits
Phosphatic
Rocks
4
Loss to Sediments
Mechanical and Autolytic Release of Phosphate
organic phosphorous
inorganic
— phosphorous
plant nutrients
Figure 6. The Biogeochemical Cycle of Phosphorus -- After
Russel-Hunter (1970).
27
or algal, bridges the deficiency and produces the nitrogen
necessary for the synthesis of algal protein. In the
absence of plentiful supplies of phosphorus, however,
nitrogen fixation was found to be unimportant.
Thus, for many years the key importance of phosphorus to the growth of aquatic algae was taken as an
absolute fact, and indeed the majority of water chemists
and limnologists never did doubt that fact and do not do so
now.
The first hints of the furor yet to come appeared
in 1967 when Willy Lange, a chemist turned botanist at the
University of Cincinnati, published a paper entitled
"Effect of Carbohydrates in the Symbiotic Growth of
Planktonic Blue Green Algae with Bacteria." His thesis
showed that algae always exist in association with bacteria
and that the association is mutually supportive. That is,
the algae utilize carbon dioxide and sunlight to produce
organic matter and oxygen by photosynthesis and the bacteria
use oxygen in the decomposition of organic matter to produce
carbon dioxide. Lange 's experiments proved to his satisfaction that it was the presence of large amounts of organic
material in water that made the production of huge amounts
of carbon dioxide available for algal growth.
Then L. E. Kuentzel, a physical chemist, in 1969
after reviewing the literature on eutrophication, concluded that carbon, not phosphorus, was the element that
28
controls algal growth. Kuentzel felt that only bacterial
action on organic matter could produce the amount of carbon
dioxide required for rapid growth. He showed that in many
cases of excessive growth, dissolved phosphorus was
exceedingly small, in fact, lower than the levels presented
by Sawyer.
In the 1970 Purdue Industrial Waste Conference, Pat
Kerr, a plant physiologist at the Federal Water Quality
Administration (FWQA), now EPA, Southeast Water Laboratory,
presented results where she concluded that carbon was the
controlling element.
Both the phosphorus and carbon schools agree that
algae need, for growth, sources of inorganic carbon, phosphorus, nitrogen, and numerous other elements such as
micronutrients. Both schools agree that algae and bacteria
generally coexist, and the phosphorus school is willing to
concede that the relationship may be symbiotic. But on
almost all other points, they disagree (Table 3). This
disagreement emanates from two basic areas of contention:
1.
Precisely how much phosphorus do algae need for
excessive growth?
2.
What sources of carbon are available to algae?
The carbon school maintains that only very small
amounts of phosphorus are needed
l
it points to the low dis-
solved phosphorus concentrations found in the water of
29
Table 3. Comparison of the Two Schools of Thought
Carbon-is-key school
believes:
Phosphorus-is-key school
believes:
Carbon controls algal
growth.
Phosphorus controls algal
growth.
Phosphorus is recycled
again and again during and
after each bloom.
Recycling is inefficient:
some of phosphorus is lost
in bottom sediments.
Phosphorus in sediment is a
vast reservoir always
available to stimulate
growth.
Sediments are sinks for
phosphorus, not sources.
Massive blooms can occur
even when dissolved
phosphorus concentration is
low.
Phosphorus concentrations
are low during massive
blooms because phosphorus
is in algal cells, not
water.
When large supplies of CO2
and bicarbonate are present,
very small amounts of
phosphorus cause growth.
No matter how much CO 2 is
present, a certain minimum
amount of phosphorus is
needed for growth.
CO 2 supplied by the
bacterial decomposition of
organic matter is the key
source of carbon for algal
growth.
CO2 produced by bacteria
may be used in algal
growth, but main supply is
from dissociation of
bicarbonates.
By and large, severe reduction in phosphorus discharges will not result in
reduced algal growth.
Reduction in phosphorus
discharges will materially
curtail algal growth.
30
eutrophic lakes during algal blooms and believes that
nutrients, including phosphorus, are recycled by organisms
during growth and released for reuse during the periodic
dieoff periods. On the other hand, the phosphorus school
takes the position that algae require relatively substantial amounts of phosphorus and the incidence of low
dissolved phosphorus was due to the uptake by the algal
cells.
The carbon school believes that the availability
of utilizable carbon is the key and that diffusional processes are too slow to permit atmospheric CO 2 to support
massive growth, hence its interpretation of the importance
of bacteria-produced CO 2 . The phosphorus school points to
the fact that algae can use, in addition to free CO
2'
carbon
dioxide produced by the dissociation of dissolved bicarbonates. Phosphorus supporters say that the dissociation
occurs so rapidly that supply of carbon dioxide cannot
possibly be limiting, and they ridiculed the carbon school
emphasis on the need for respiratory supply.
Blue green algae are an index of eutrophication.
They are able to develop under adverse pollution conditions,
i.e., high salinity. But, by the time they begin to grow,
inorganic nitrogen is often scarce and P is sometimes
almost undetectable as a result of the nutrient uptake by
the existing plankton. This paradox hardly supports the
hypothesis that N and P are the only important components
31
responsible for the change in flora induced by eutrophication. Provasoli (1969) concluded that other factors must
be at work; he showed that blue-greens are the only freshwater algae that show an affinity for Na and K. Since Na
and K are found in excess in domestic sewage we could
hypothesize that these elements control the species development. Many arguments have been made about the effects
of Ca, Mg, Cl, and SO 4 values in the Great Lakes. However,
no functional relationships have been reported. The indispensability of silicon for diatoms is well documented
(Lund, 1965). The need for iron, manganese, boron,
molybdenum, zinc, cobalt, copper, and vanadium has been
recorded for a few single algae (Wiessner, 1962), but we
cannot infer that all algae need them, nor do we know
whether certain species, genera, or algal groups have
peculiarly high requirements for any of them.
Further contributions to the nutrient core result
from biological activity. Some zooplankton are omnivorous
while others are herbivorous or carnivorous. Regardless,
a certain percentage of the phytoplankton which is eaten
is anabolized and certain portion is excreted (Figure 4,
5
).
All aquatic systems have a certain amount of dead
and decaying organic matter called detritus. The detritus
is decomposed by the bacteria (Figure 4, 1 14) reasing
nutrients (Figure 4, 1 3 ). The insensitive Biochemical
32
Oxygen Demand (BOD) technique is a measure of the existing
biologically oxidizable detritus. However, there are more
sensitive tests for organic matter.
Acting on the organic matter in the water are the
bacteria. Bacteria feed on the organic matter (Figure 4,
1 13 ) and break it into inorganic elements (Figure 4, 1 4 ).
The bacteria create the oxygen demand through respiration.
Different species of bacteria are capable of acting at
different rates and liberating different elements from the
organic matter. The needs of the bacteria (anabolism)
differ with the species. But, the nutrient contributions
of bacteria are poorly understood.
Certain nutrients are required for phytoplankton
growth (Figure 4, 1 8 ) It is well known that certain algae
prefer specific nutrients (i.e., Cyanophyta have an
affinity for high sodium and potassium levels). We also
recognize that some algae are "luxury" consumers of certain
nutrients. These algae have the ability to store large
amounts of nutrient (i.e., phosphate) while only using a
part of it.
The life cycle of algae varies with the species
involved (Figure 7). Algae have been shown to reproduce
in a 24 hour interval with massive blooms occurring in one
week. The death rate depends upon the species of algae
involved. This death rate can be natural (Figure 4 ,15 )
33
Figure
7. Simple Motile Green Alga, Chlamydomonas --
After Cockrum, McCauley, and Young gren (1966).
34
or accelerated because of eutrophication or toxic effects
(Figure 4, 1 7 ).
Each year a large amount of the nutrients are lost
to the sediment (Figure 4, 1 2 ). The sediment loss is due
to many factors:
1.
Nutrients can precipitate out and fall by gravity
to the benthos.
2.
Sediment in the system can adsorb nutrient and
leach them out as the particles settle.
3. A portion of the phytoplankton and zooplankton that
have died settle also into the sediment and their
nutrient value is lost.
The Phytoplankton System
Phytoplankton are the microscopic wandering plants
of the aquatic system. They are controlled largely by the
incoming solar radiation (Figure 4, 1 9 ). The radiation is
necessary for the chlorophyll in the plants to carry on
the process of photosynthesis. The saturated or optimal
growth rate of phytoplankton (Figure 8) has been shown as
a function of the solar radiation. The growth rate increases with solar radiation up to about .1 langleys/min.
Above this intensity the growth rate is partially inhibited.
As the solar radiation passes through the water,
two changes take place. The intensity of light is reduced
and the quality of the light is changed (Figure 9). This
35
25°
2.0
LU
C-1
0
.
10
SOLAR RADIATION
. 20
.30
(LANGLEY/MIN.)
Figure 8. Phytoplankton Growth Rate as a Function of
Incident Solar Radiation Intensity and
Temperature -- After DiToro et al. (1970).
36
PERCENTAGE OF SURFACE LIGHT
5
2
10
20
50
60
70
525
80
90
75
100
Figure 9. Depth to Which Light of Various Wave-Lengths
Penetrates, Expressed as Percentage of the
Surface Light -- After Rabinowitch and
Govindjee (1969).
37
is a result of the absorption of the blue-green wavelengths near the surface. The light extinction (Figure 4,
I ll ) is controlled by sediment, thermal density layers,
biological reflection, etc. It is important that the
quality of the light penetration not be affected by
impurities (i.e., sediment) since the phytoplankton absorb
light at band-limited wave lengths (Figure 10).
The temperature of the water affects the respiration
rate of the phytoplankton (Figure 4, 1 10 ). Temperature
investigations could become very complex if we were attempting to model solubilities and decomposition rates, etc.
However, at this stage, water temperature is treated very
generally.
Feeding on the phytoplankton are the herbivorous
zooplankton (Figure 4, 1 16 ). The zooplankton however, feed
at different rates on the phytoplankton population (Table
4). Some species of phytoplankton such as filamentous
blue-greens are not grazed by zooplankton. The limnological understanding of the preference of zooplankters
for various phytoplankton is very vague.
The number of phytoplankton are reduced by natural
and toxic mortality rates (Figure 4, 1 15 ) as discussed
earlier.
As the number of phytoplankton increase the biomass
begins to block out the incoming light. This self-shading
(Figure 4, 1 12 ) reduces the available light for
38
(b)
Phyco- erythrin 400
Phycocyanin
5 00600
70 0
Wavelength, nm
(c)
Figure 10. Absorption Spectra of Three Types of
Chloroplast Pigments: (a) Chlorophylls;
(b) Carotenoids; (c) Phycoerythrins and
Phycocyanins -- After Rabinowitch and
Govindjee (1969).
39
Table 4. Grazing
Rates of Zooplankton -- After DiToro
et al. (1970).
Organism
Grazing Rate
(1/mg. dry wt.-day)
Rotifer
Brachionus calyciflorus
0.6
- 1.5
Copepod
Calanus sp.
0.67 - 2.0
Calanus finmarchicus
0.05
Rhincalanus nasutus
0.3
Centropages hamatus
0.67 - 1.6
- 2.2
Cladoc era
Daphnia sp.
0.2
- 1.6
Daphnia magna
0.2
- 0.3
0.8
- 1.10
Natural Association
Georges Bank
4o
photosynthesis and reduces the photic zone.
The extreme
case may be illustrated by the water hyacinth problem
in
Florida where the surface of the water is completely
covered by these macrophytes.
As the phytoplankton respire they require a certain
amount of energy for catabolizing. This reduces the amount
of growth through anabolism during the day (Figure 4, 1 7 ).
Since plankton are wandering species and are
subject to currents and gravity, they begin to settle out
(Figure 4,
15
). The number of phytoplankton in the photic
zone are reduced through settling.
The Zoo plankton System
The zooplankton diversity is present because of the
existing phytoplankton, chemistry, and hydrodynamics of the
system. Since zooplankton can be omnivorous it is very
difficult to determine upon what they are feeding. Generally, however, the zooplankton are taken to graze upon the
phytoplankton for their growth food source (Figure 4, 1 16 ).
The preference of zooplankton for food sources was discussed earlier.
As the zooplankton catabolize (Figure 4, 1 19 ) they
reduce their growth rate by a fraction that is primarily a
function of temperature. It has been. shown by Carlson
(1968) that phytoplankton grow best when they are grazed
by zooplankton. There is an optimum zooplankton population
41
• that grazes at a rate permitting
the optimum growth rate
of phytoplankton for the existing physical
and chemical
conditions.
A certain percentage of the zooplankton must die
(Figure 4, 1 18)
Independent of this natural mortality is
the reduction of zooplankton by the small filter feeding
fish (Figure 4, 1 20 ), (i.e., shad, cisco, alewife, etc.).
It is very difficult to determine if the zooplankton
population is a result of the availability of phytoplankton
or the feeding of small fish, or both to varying degrees.
Although the zooplankton feed on phytoplankton, a
portion is utilized for zooplankton growth, the rest is
excreted (Figure 4, 1 19 ).
The zooplankton population consists mainly of
Copepods, Crustaceans, and Rotifers, and each has a different life cycle. The growth stages, separated by molts in
the life cycle of Copepods, are similar in all forms. The
egg hatches as a typical nauplius larva. After six successive molts a metanauplius form gives rise to a Copepodite
stage. There are usually five molts after the metanauplius to the Copepodite stage. The final molt from the
fifth Copepodite stage results in adult males and females.
Copepods have been shown to have one to seven generations
in a year depending primarily upon the temperature of the
water. Little is known about the longevity in Copepods,
the littoral-benthic and pond species can live for nine
42
months in the laboratory and possibly
longer in nature.
The smaller Copepods live one to
six months and in some
cases may last two or three years (Hutchinson, 1967).
The
life cycle of the zooplankton
is very complex with fish
exercising a considerable control on the nature of zooplankton associations.
Attempts at modeling rotifers (Figure 11) are still
in the infant stage. We generally say that the zooplankton
have a constant death rate. The zooplankters can die
because of natural mortality (Figure 4, 1 6 ), toxic effects
through phytoplankton (Figure 4, 1 17 ), or grazing by
larger carnivores (Figure 4, 120).
43
0
4-3
C:r1
1-
(1.)
4-
-0
0
0
4-
t-
r()
CHAPTER 3
THE STUDY AREA
Locations and Water Quality
Considerations
Lake Mead, the largest reservoir in the Western
Hemisphere, is located on the Colorado River (Figure 12).
The impount is a canyon-type reservoir formed by Hoover
Dam, a concrete arch gravity structure that has a maximum
height of 726.4 feet. Although the Dam was completed in
1936, storage began during the previous year. Below the
Dam, the tailrace releases show a weak seasonal oscillation
in salinity (Figure 13). At full elevation, the reservoir
extends a distance of 115 miles upstream and has a capacity
of 31,047,000 acre-feet. A perimeter of 550 miles encompasses 158,000 surface acres with a maximum depth of 589
feet. The Lake is extremely irregular in shape. Boulder
and Virgin Basins (Figure 14) contain about 60 per cent of
the total storage in the reservoir.
The climate at Lake Mead is arid. Mean annual
temperature at Las Vegas is 66°F (19°C) and mean annual
precipitation is less than 5 inches, according to Weather
Bureau Records. Maximum temperatures of 110°f (43°C) are
non uncommon in July and August. Average minimum temperature in January is 30 ° F (-1 ° C). Winds are generally light.
44
115
Horseshoe Reservoir
Bartlett Reservoir
Roosevelt Lake
Apache Lake
Canyon Lake
Saguaro Lake
Figure 12. Map Showing the Colorado River Basin from
Lake Powell to the Mexican Border
46
3
3
CO'DI
oyoac
0.00,1;
VO'c8CUCD'uo
( 0:XNJd NI .L.IINPHS
12,312
47
36 ° 35'
Basins
I. Boulder Basin
2. Virgin Basin
3. Temple Bar Area
and Virgin Canyon
4. Gress.Basin
5. Grand Wash Bay
6. Pierce Basin
7. Lower Granite Gorge
8 Overton Arm
30'
25'
,
ii4° 501
451
40'
35'
30'
25'
20'
15'
10'
05'
11t, ° 00
Figure 14. Map of Lake Mead Showing Basins
1
57' 55'
48
The water in Lake Mead is derived from
one major
source, the Colorado River, and three minor sources, the
Muddy and Virgin Rivers and Las Vegas Wash. Earlier
reports by the EPA (1968a) indicated that the upper reaches
of Lake Mead had a limited algal population because sediments cause reduction of light penetration and adherence
of particles to the algal cells, thus speeding their
settling to the bottom. Nutrient contribution of the
Colorado River to the upper reaches of Lake Mead is of
concern.
The Colorado River contributes a mean April discharge of 15,200 cfs for 98 per cent of the total flow into
Lake Mead. A seasonal oscillation in salinity at the Grand
Canyon Station shows considerable watershed effect (Figure
15).
The Virgin River flowing into the northern end of
the Overton Arm is characterized by large amounts of silt
transported from the watershed. Although the mean April
flow is about 246 cfs, the Virgin River contributes only
1.5 per cent of the flow and 2 per cent of the solublephosphorus to Lake Mead. The yearly cycle of salinity
increase indicates the contributions of salts from surface
runoff (Figurc 16).
The Muddy River flowing into the northwest end of
the Overton Arm behaves much like the Virgin River on a
lesser scale. The mean April flow of 38 cfs accounts for
49
-
-
cr,
I—
I
CO*01, 1,
CO'OZI
C:OCOl
:LI
COO:9
CO•CS
NI
Cf:'
I
(
50
I
—
cr)
n
27
7
z
-
•
8
fz.!
-
----/
__
—
CP N
1-4-0000Z
4.---100'0a-e.
t
1
0000i.
00'0.1t
i
OIX 111,1,1 NI
I-
OirCZI
)
LI INFItiS
i
I:0'U
1----,
CWO,
i
8. —
owe
.
51
one per cent of the total phosphorus and .3 per
cent of the
flow into Lake Mead.
Las Vegas Wash contributes about 35 per cent of the
total soluble phosphorus to Lake Mead. The Wash has a mean
April flow of 31 cfs and accounts for .2 per cent of the
water input to the Lake. The discharge from Las Vegas Wash
shows a definite seasonal oscillation with the highest
flows occurring in the winter. One of the reasons for
the increase in flow is the seasonal rainfall. The major
reason is the influx of tourists who come to Las Vegas for
the warm sunny winters. The 1971 report by Everett on the
cause-effect relationship of Las Vegas shows a strong increasing trend in the discharge from Las Vegas Wash for the
years 1959 to 1969.
Recreational Uses
The Lake Mead Recreation Area is a popular watersports area. Much of the human influx is from touristoriented Las Vegas, Nevada. All commercial channels are
being explored to bring Las Vegas guests to the reservoir.
Although sightseeing is the most popular visitor activity,
swimming, boating, water-skiing, and fishing are an important part of the recreational use. The Bureau of Reclamation provides guided tours through Hoover Dam. Over 615,000
visitors inspected the Dam in 1969.
52
The National Park Service (NPS), Boulder City, is
responsible for the administration of the Lake Mead
Recreation Area. The NPS has provided fourteen boatlaunching ramps, two supervised swimming beaches, and has
twelve concessions operating in the area. There are
seventeen campgrounds and twenty-four picnic shelters
around the Lake. Visitor use to the Lake Mead National
Recreation Area in 1969 was over 6 million (Hoffman and
Jonez, 1971).
Las Vegas Bay is heavily used for water-based
recreation, including water contact sports. A marina is
located on the Bay near the mouth of Las Vegas Wash. Excessive algal growths are causing distinct green color,
odors, and nuisance conditions.
CHAPTER
4
EXPERIMENTAL DESIGN AND METHODOLOGY
State Estimation
The investigation at Lake Mead was directed toward
providing some insight into a number of practical and
academic problems. The simplest problem was to establish
a station at Las Vegas Bay to monitor chemical, biological,
and some hydrodynamic processes to provide a baseline for
in-progress fish studies directed by Dr. J. Deacon, University of Nevada at Las Vegas.
The second problem was to determine the relative
behavior of different parts of Lake Mead to quantify the
extent of existing pollution in the reservoir. This was
done by distributing the sample locations over the reservoir.
The third and most complex thrust of the study was
to characterize the system through an analysis of state
sets (Figure 17). To establish a state set of the system
we had to quantitatively measure all of the "important"
variables (chemical, biological, and hydrodynamic) at one
time, at one depth, and at one location. Each state set
was in fact an "in situ" bioassay. This insured that the
change of state and the biological response had occurred
under natui
conditions. An intensive literature search
53
54
STATE SETS
Class I:
Variables
30
Depths
2
Times
6
Locations
8
Number of State Sets in Class I: 2 x 6 x 8
Class 11:
Variables
14
Depths
8
Times
6
Locations
8
Number of State Sets in Class II: 8 x 6 x 8
Class III:
Variables
11
Depths
10
Times
6
Locations
8
Number of State Sets in Class III:
Class IV:
Variables
Depths
96
=
10 x 6 x 8
384
-
480
2
31
Times
6
Locations
8
Number of State Sets in Class IV:
31 x 6 x 8
=
1488
Figure 17. Four Classes of State Sets Investigated
55
resulted in a list of parameters that would describe
the
state of the system. Financial limitations resulted in the
measurement of 30 independent variables, and
3 dependent
variables. The dependent variables include: (1) time, (2)
space, and (3) depth.
The independent variables (Figure 18) include
biological, chemical, and physical parameters. The tradeoffs associated with the sampling design in the Lake Mead
investigation are typical. The advantages of the sampling
method include:
(1) reduced cost, (2) greater speed, (3)
greater scope, and
(4) greater accuracy.
The sampling theory suggests that it costs less to
sample a small fraction of the aggregate. Sample data are
inherently from a smaller volume and can result in information more quickly. A larger problem (scope) can be handled
by small flexible sample programs. Since personnel of
higher training can be directly in control of reduced work
volume the accuracy of the results may be improved.
The steps recommended by Cochran (1963) in a sample
survey should incluGa: (1) objectives of the survey, (2)
population to be sampled, (3) data to be collected,
(4)
degree of precision desired, (5) methods of measurement,
(6) the frame (list of sampling units), (7) selection of
the sample,
(8) the pretest, (9) organization of the field
work, (10) summary and analysis of data, and (11) informasampling design in the
tion gained for future surveys. The
56
Phytoplankton
14C Primdry Productivity Rate
Asterionella
Ceralium
Biological
Daphnia
ina
roratella
Bossu
Zooplankton
Ni
Calanoid
Cyclopod
Fe
Mn
__ Micro
Zn
Cu
SO
4
CO3
Nutrient-
NO3
PO
__ Macro
4
Mg
Ca
Chemical
Na
pH
Environwent
Dissolved Oxygen
Soluble Salts
Electrical Conductivity
Cl
Physical Temperature
Solar Radiation
Light Extinction
Figure 18. Independent Variables Investigated
57
Lake Mead investigation closely followed the steps listed
above. Of greatest value in the list was the pretest
carried out on Lake Mead in June, 1969. The chemical and
biological results of the brief feasibility study not only
provided insight into the changing parameters but also
provided information for steps 2 through 9.
There are two approaches in a systematic sample
design program. The first approach is an ex ante design,
before the fact, based upon past and current knowledge of
the system. The second approach involves an ex post design,
after the fact, in which the results of a completed investigation can be used to evaluate the original sample design.
Too often the ex post evaluation of the sample design is
neglected at the end of an investigation.
Of major concern in the sample survey was the data
network density. Since the feasibility study indicated
considerable changes across the Lake, inputs to each of
the basins of the Lake had to be monitored. Since all
costs of transportation were absorbed by the Bureau of
Reclamation, the major criterion for the sample site location was ex ante data provided by subjective attitudes that
were modified by the feasibility study.
There was no compromise between the data network
The spatial
density and frequency of temporal sampling.
distribution was foremost in our criteria. It was assumed
that temporal changes would uniformly alter the magnitudes
58
of the ppr over the entire set of spatial sampling points.
Since the time of thermal stratification and degeneration
was not known, we were satisfied to sample once during the
summer stable period and once during the winter instability
period. The other sample dates were determined primarily
by cost and convenience of those participating (i.e.,
holidays). The hydrodynamics of the Lake were (and are)
poorly understood and so did not generally influence the
exact times of sampling. More studies, however, are necessary in this direction.
Most investigators including Patrick (1971) concluded that laboratory manipulation of parameters introduces error. Patrick's work with the manipulation of light
and temperature concluded that artificial light does not
bring about as great an increase in diatom development as
natural light. Patrick felt that optimum conditions for
light and temperature form a fairly narrow range within
the range of tolerance. Increases in temperature near the
lower end of the range of tolerance improved the structure
of the diatom community significantly while increases near
the upper end of the range of tolerance produced severe
degradation in community structure. Sampling locations
were selected using the criteria of maximizing the number
of sets of states. In other words, to maximize the interpretive quality of the data. The experimental design is
59
given in Table 5. The locations of
sampling are shown on
the map (Figure 19).
Sampling Dates
Field investigations were made from the summer of
1970 to the winter of 1972. The survey dates and season
of the year are given in Table 6.
By scheduling the surveys as described above, data
were collected during periods of the Lake's annual temperature cycle. The June and September surveys were indicative
of thermal stratification while the other dates were designed to note the effects of winter and spring mixing.
Sampling and Measurements
Each of the eight water-quality stations was sampled
identically. All water samples for analysis were collected
using three- and six-liter, polyvinyl-chloride (PVC) Van
Dorn samplers. The PVC sampler precludes any ion exchange.
Methods of sampling and measurements are as follows:
1.
Continuous temperature profiles to 54 meters were
obtained using a Precision Scientific temperature
probe.
2.
The DO continuous readings were taken from a
galvanic-cell oxygen analyzer manufactured by
Precision Scientific. The meter was calibrated
each day using the Azide-Modification of the
6o
•
Cln
H
1.)
;-I
b.0
0
H
2-, S-n
EC)
0 rgl
F-,
0
+ E
rz-i
44
a) <
ai
(1.)
,...0
0c-N
H
cn
I
0
k....0
0
.—I
e•-n
O
,..0
g
•ri
...
H
Cl
01
0
P
,. .D
O
0
If \
tr
0
L4 III
0
v-N.
H
0
H
c•-%
I
0
‘..D
H
a-,
,..0
H
\
0
0
,..0
r.
Lr1 0
\
8
61
62
Table
6. Survey Dates and Seasons
Season
Dates
Summer
September 6-11, 1970
Fall
November 24-29, 1970
Winter
January 23-27, 1971
Winter
February 25-27, 1971
Spring,
April 3-8, 1971
Summer
June
Winter
January
4-8, 1971
8 -1 3
, 1 97 2
Winkler technique as presented by Standard Methods
(American Public Health Association, 1971).
3.
A portable Beckman Electremate pH meter was used to
' obtain the hydrogen ion levels. The pH meter was
also used to obtain the alkalinity values as described by Standard Methods (American Public Health
Association, 1971).
4.
Chlorophyll samples at five and thirty meters were
taken. Two milligrams of magnesium sulphate were
added to 1,000 ml of water and filtered through a
.45-micron membrane filter.
5. Complete chemical analysis (Figure 18) was done on
dark
a 1-liter sample. The samples were kept in
63
glass bottles which had been washed with a dilute
acid solution. Since micro-elemental analysis was
also done, no preservatives were added. Samples
for chemical analysis were taken at five and thirty
meters.
6.
Phytoplankton and zooplankton samples were taken at
0, 1, 3, 5, 7, 10, 15, 20, 25, 30, and 35 meters.
The phytoplankton sample was placed in a six-dram
vial and preserved with Lugol's reagent. Each of
the zooplankton samples were concentrated from a
six-liter sample to six drams using a fine mesh. A
ten per cent formalin solution was used to preserve
the zooplankton.
7.
Light transparency was measured with a photoelectric cell rather than a Secchi disc. The cell
gives considerably more accurate results and is
corrected for changes in solar radiation. Also,
the cell affords the opportunity to locate turbidity
layers. Since this study was primarily concerned
with the photic zone, we required an accurate light
meter to determine the compensation depth. The
compensation level is the depth at which one per
cent of the incoming solar radiation can be measured.
8. Solar radiation was measured with a Belfort
recording pyrheliometer that was calibrated each
day with an Eppley pyranometer.
64
Primary Productivit y__ 14 C Method
To determine the total algal production in a given
water column, it is necessary to run a vertical series of
measurements with samples from various depths. Each depth
is subject to different light, temperature, etc. conditions
and is indicative of a separate state. Water samples were
taken at 0, 1, 3, 5, 7, 10, 15, and 20 meter depths using
a 3-liter PVC Van Dorn sampler. The PVC sampler precluded
any contact with bare metal which may be either detrimental
to the algae (Doty and Ouri, 1958) or stimulating (Goldman,
1963). The water sample at each depth was divided into
three containers: 125 ml light and dark bottles and a 500
ml sample for alkalinity and pH values. The transparent
light bottle permitted the incoming solar energy to act on
the chlorophyll, thus permitting photosynthesis. The dark
bottle excluded all penetrating light. Care was taken not
to expose water samples taken at any depth to the surface
solar radiation since the intensity creates light injury
to the phytoplankton. To protect the phytoplankton from
covers
light shock, the light bottles were placed in black
during preparation periods and were kept in wooden boxes
exterior.
painted black on the interior and white on the
The principle used in the
14 C technique is the in-
14 C) in the organic matter of
corporation of a tracer (
a measure of
phytoplankton during photosynthesis which is
the rate of primary production.
If the content of total
65
CO
of the experimental water is known
and if a definite
ItI C
amount of
is added to the water, then by determining
2
the content of 14 C in the plankton after the
experiment
the total amount of carbon assimilated can be calculated
(Vollenweider, 1971).
From the alkalinity and pH determination the amount
of C
.5
in the water is found. To the light and dark bottle
micro-curies of 14 C was added to the bottom of each
12
bottle with a hypodermic needle and long cannula. The
bottles were shaken by hand, stoppered, and returned via a
calibrated line to the same depth, light, temperature, etc.
condition for the incubation period.
The bottles were placed in the water as close to
10:00 a.m. as possible and removed after a four hour
interval. The experimental bottles were withdrawn from
the various depths and stored in a black box until the
beginning of the filtration operation. The bottles were
removed as quickly as possible from the water to reduce
light energy and further photosynthesis. Fifty ml aliquots
were then transferred through a filtration apparatus from
the light and dark bottles onto 0.45 micron Millipore
membranes. The samples were filtered as soon as possible
after removal from the water. The filtration apparatus
hand pump. Care was
was connected to a negative pressure
and aspiration
taken to rinse the 50 ml graduate cylinder
funnel to insure that all radioactivity passed onto the
66
membrane. After removal of the filters from
the filtration
unit they were placed onto waxed paper and stored until
returned to the University.
CHAPTER
5
RESULTS AND DISCUSSION
Data Description
Up to the commencement of this study, available
data were collected without coordination. It is fragmentary
and of no utility for diagnostic purposes. Therefore, there
was no quantitative understanding of the behavior of the
systems and the stages leading to present conditions. The
reasons for the limited value of previous investigations
are: (1) the restricted spatial sampling, (2) the limited
temporal sampling, and (3) the lack of essential information for identifying eutrophication trends.
Temperature
The annual temperature regime in Lake Mead dewater
scribes a warm monomictic cycle. These lakes have a
temperature which is never below 4 ° G and freely circulate
typical
during the winter at or above 4°C. One of the
temperature regimes is presented in Figure 20 for the
Beacon Island station.
in Lake
Thermal stratification is well documented
solar radiation warms the
Mead. In the summer months the
The warm water is less
surface waters of the reservoir.
67
68
CO
c•I
csi
CD
0.00
6.00
16.01
21.01
32.01
DEPTH IN METERS
40.01
48.02
69
dense than the cooler bottom waters resulting in a thermal
density layer.
Since water is a very poor conductor of heat, we
generally observe a summer temperature profile as seen in
Figure 21. The discontinuity layer is called the thermocline. It is characterized by a drop of 1°C in temperature for each depth increase of one meter. The upper warm
layer of water is referred to as the epilimnion. The dense
cooler waters below the thermocline are called the hypolimnion.
The temperature range was 10.1 to 28.4°C, with
November, January, and February temperature profiles showing
the isothermous nature of the Lake during winter turnover.
It is at this time that nutrients released in the anaerobic
hypolimnetic layers of summer are brought to the surface.
The homogeneous winter temperatures are 11 + 1°C. The June
figures indicate the beginning of the thermocline. In
September we can distinguish between the epilimnion, thermostation. The thermocline in
cline, and hypolimnion at each
and had a
September was found between 18 and 28 meters
temperature range of 17 to 26°C.
Dissolved Oxygen (DO)
cycle in Lake Mead can be
The dissolved oxygen
(Figure 21). In
described as a negative heterograde scheme
with
this ease we observe a minimum DO in the thermocline
70
25
50
°
Dissolved Oxygen (ppm) - Temperature ( C)
Location: Bureau Raft
and
Figure 21. September and January Dissolved Oxygen
Temperature Profiles
71
higher saturations in the epilimnion and hypolimnion. The
DO pattern is similar at all stations during the seasonal
changes. The range of DO is 1.8 to 13.0 ppm. The November,
January, and February DO isopleths indicate that the winter
turnover is reaerating the system. The September charts,
however, show that the Lake goes into stress in the hypolimnion during summer stratification. DO values of two ppm
at the thermocline depth indicate a high Biochemical Oxygen
Demand (BOD). Since the DO level recovers from this zone
of high oxygen demand we postulate that high levels of
organic matter are oxidized at the thermocline level. As
the algae sink under gravity they meet the denser cold
waters capped by the thermocline. Here bacteria break the
algae down, releasing ions, and using up DO through respiration. The DO levels in the hypolimnion in September are
considerably below the recommended level for cold-water
fish regeneration.
Soluble Salts
The September data (Figure 22) indicate across all
the stations that with a well-developed thermocline we
salts at 30 meters
observe the highest values for soluble
a 100 ppm inin the hypolimnion. There is approximately
Cove to the Bureau raft
crease in soluble salts from South
turnover the 5at all seasons of the year. During winter
The range of soluble
and 30-meter values are very cic
,
72
87 0
770
6 70
A
DEE
A
B
C
D
G
H
BUREAU RAFT
VEGAS BAY
BEACON ISLAND
BONELLI LANDING
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
Figure 22. September and January Soluble Salt Concentrations
73
salts is from 600 to
84o ppm. The soluble salts in
Vegas Bay do not appear to seriously increase the level
at the raft. Higher jumps in soluble salts are noted
between South Cove and Temple Bar.
Electrical Conductivity (EC)
EC does not appear to be affected by thermal
stratification (Figure 23). There is a conductivity range
of about 0.20 x 10 3 micromhos at each station throughout
the year. The EC ranges between 0.95 and 1.25 x 10 3
micromhos over the eight stations. The specific conductance
of an oligotrophic lake is less than 200 micromhos at 18°C.
These levels of specific conductance place Lake Mead deep
into the range of a eutrophic lake. Conductivity, however,
is a very weak index of eutrophication.
H dro en Ion Concentration ( H)
The pH generally decreases slightly across the
basins (Figure 24). During summer stratification the lower
pH values were consistently found in the hyplolimnion. The
pH ranged between 7.8 and
8.4. We noted that the Las Vegas
levels
Wash station does not appear to be affecting the pH
as to the
in Boulder Basin. There has been some debate
in Boulder
industrial effect on the pH of the groundwater
Basin.
74
23-29 Jan 1971
1.3 0
1.20
.90
A
B
C
D
BUREAU RAFT
VEGAS BAY
BEACON ISLAND
BONELLI LANDING
E
F
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
Figure 23. January and September Values of Electrical
Conductivity
75
=Q-
AB
C
DEE
A
B
C
D
G
BUREAU RAFT
VEGAS BAY
BEACON ISLAND
BONELLI LANDING
H
E
F
MINER'S COVE
ECHO BAY
G
H
TEMPLE BAR
SOUTH COVE
Figure 24. September and January pH Changes
76
Chloride, Magnesium, Potassium,
Sodium, and Calcium
Each of these elements appear to increase across
the basins. This is primarily a result of salt loading
and salt concentration. All of these elements except Mg
appear to have much lower September values in South Cove.
However, the levels of these element nutrients are very
high. The level of chloride is five times as high as found
in Lake Erie (Figure 25). The calcium levels show a strong
response to thermal stratification (Figure 26).
Phosphate
The ortho-phosphate levels found in Lake Mead fall
in the range of an oligotrophic lake (0.1 to 0.3 ppm).
Boulder Basin appeared to be richer in PO 4 than Virgin
Basin (Figure 27). The PO4 levels at South Cove drop across
Gregg's Basin. This could be a result of complexing and
sedimentation or increased primary productivity. The
minimum ortho-phosphate level for excessive crops of algae
is equal to or greater than 0.01 ppm (Lee, 1970).
Nitrate
The nitrate values are much higher in September
and November. There is no increase or decrease across
the system; therefore, the Colorado River is responsible
for introducing these high levels (Figure 28). The minimal
bloom must be equal to
nitrogen content for excessive algae
77
120
110
100
go
80
A
DEF
G
H
23-29 Jan 1971
5 meters
120
30 meters
110
loo
go
80
DEF
AB
A
B
C
D
G
BUREAU RAFT
VEGAS BAY
BEACON ISLAND
BONELLI LANDING
H
E
F
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
Figure 25. September and January Chloride Concentrations
78
4-11 Sept 1970
.0 4.
90
o
so
70
E
60
50
DEE
A
G
H
25-28 Nov 1971
5 meters
— — — 30 meters
A
B
C
D
Figure
BUREAU RAFT
VEGAS BAY
BEACON ISLAND
BONELLI LANDING
26. September and
E
F
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
November Calcium Concentrations
79
0. 05
0.04
0.03
0.02
ci0.01
0.00
A
B
C
D
BUREAU RAFT
VEGAS BAY
BEACON ISLAND
BONELLI LANDING
E
F
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
Phosphate Levels (PO 4 )
Figure 27. September and April
80
A
B
C
D
Figure 28
.
BUREAU RAFT
VEGAS BAY
BEACON ISLAND
BONELL I LANDING
E
F
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
September and April Nitrate Concentrations
81
or greater than 0.3 ppm (Lee, 1970). Excessive nitrogen
is available at all times of the year. Boulder Basin does
not contribute excessive nitrate values.
Bicarbonate
The bicarbonate (HCO
3
)
values show excellent re-
sponse to thermal stratification in September (Figure 29).
The higher values are found 30 m beneath the surface. The
trend is toward reduced HCO 3 values across the Lake. The
summer levels are 10-15 ppm less than the cool winter
turnover values. Excessive HCO
3'
based on ppm data, is
available throughout the year. The carbonate levels read
zero at all times of the year.
Iron, Manganese, Zinc
These elements do not show a consistent trend
across the system. Manganese is available in the optimal
range. The iron and zinc levels do not appear to be
limiting. The highest zinc values were repeatedly recorded
of microat Beacon Island (Figure 30). Although the levels
in the
nutrient requirements in not known, their presence
their availability.
ppm range is taken as evidence of
Copper
not appear to follow any trend
The copper levels do
(Figure 31). Generally copper is toxic to all algae at
concentrations greatthan
0.05 ppm (Hutchinson, 1967).
82
170
1 1+0
110
5
170
meters
—__ — -50 meters
1 1+0
110
A
B
C
D
BUREAU RAFT
VEGAS BAY
BEACON ISLAND
BONELL I LANDING
E
F
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
Bicarbonate Values
Figure 29. September and January
83
.10
.05
.00
A
DEF
G
H
.10
5 meters
---
30 meters
.05
.00
A
B
C
D
Figure 30.
BUREAU RAFT
VEGAS BAY
BEACON ISLAND
BONELLI LANDING
E
F
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
February and June Zinc Concentrations
84
.10
.o5
.00
.10
.o5
.00
ABC
DEE
A
B
C
D
G
BUREAU RAFT
VEGAS BAY
BEACON ISLAND BONELLI LANDING
H
E
F
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
Copper
Figure 31. February and June Concentrations of
85
This level is exceeded many times in Lake Mead. However,
the Cu level at certain times of the year is too low to
detect.
Sulphate
The sulphate levels consistently increase by more
than 60 ppm across Lake Mead (Figure 32). This is a result
of extensive chemical action in the reservoir. The sulphate begins to increase at South Cove. Part of the increase can be attributed to salt concentration. Part is a
result of the hydrobiology. In Gregg's Basin we get high
productivity resulting in a high Biochemical Oxygen Demand
(BOD) in the bottom of the Basin. This creates low dissolved oxygen conditions and low electropotential (Eh).
Under these conditions hydrogen sulphide (H 2 S) is released
and iron is precipitated out. Some of the available H 2 S
goes into solution as sulphates. The sulphate level in
Boulder Basin is 15 times as high as it is in Lake Erie.
Light Intensity
The depth of penetration appears to increase in
warm months. The one per cent level appears to be conthe
sistently deeper at Bonelli Landing and Temple Bar with
Light penetraformer being the clearest water in the Lake.
tion does not consistently improve across the system as
in Boulder Basin is
sediment is lost. The transparency
often as poor as South Cove.
86
A
B
360
5 meters
30 meters
3 00
24o
DEE
AB
A
B
C
D
G
BUREAU RAFT
VEGAS BAY
BEACON ISLAND BONELLI LANDING
H
E
F
G
H
MINER'S COVE
ECHO BAY
TEMPLE BAR
SOUTH COVE
Figure 32. September and January Sulphate Concentrations
87
Zooplankton
The important groups generally represented in the
zooplankton of lakes belong to the group of nonphotosynthetic: Protista, Rotifera, and Crustacea. In addition to
members of these groups, a few coelenterates, flatworms,
mites, and larval insects may be found in the plankton.
Lakes generally have three major distinct habitats: the benthic, which is the mud-water interface area;
the littoral, which usually consists of the shore line
habitat; and the pelagic, which is the free, open water
area. The Lake Mead investigation was restricted to the
pelagic area.
The thermocline in a lake is that area which,
during the summer months, separates the upper warm water
called the epilimnion, from the bottom colder water, referred to as the hypolimnion. When the water temperature
of the epilimnion equals that of the hypolimnion. the lake
turns over, i.e., the water in the epilimnion begins to
circulate with the water in the hypolimnion. This condition
circulates the nutrients that were trapped in the hypolimnion during the summer months and reintroduces them into
the upper layer of water. This condition persists until
of
the spring. The warming trend plus the availability
bloom. The
the nutrients cause what is known as the spring
begin
phytoplankton take advantage of these conditions and
to reproduce rapidly. This phytoplankton bloom (Figure 33)
88
r:Q
r-1
Li"
Q
cp
csi
Q
if\
o
+>
4
w 1J3d SWS I NVn0 JO )J39141111
CrN
bi)
-ri
•
89
is usually soon followed by an increase in the number of
zooplankton (Riley and Bumpus, 1946).
Other Zooplankton Studies on Lake Mead
Moffet (1943) stated that Ceratium in Lake Mead was
the most abundant organism, followed by Calanoid, Cyclopoid,
Microcystis, Daphnia, and Polyarthra (a rotifer). No quantitative data were given. Comparing this with the zooplankton of 1971 we find that Ceratium is still the most
abundant organism. But that is where the similarity ends
Comparing the organisms found in 1943 we see that in 1971
the next most abundant organisms were the Cyclopoids, and
a rotifer, Keratella. Since Keratella is the dominant
rotifer found in Lake Mead, we assume that it now occupies
the ecological niche that Polyarthra once held. In 1971,
the rotifer Polyarthra was no longer present; this could
mean that while Keratella was establishing itself in Lake
Mead, Polyarthra found it very difficult to compete.
Hutchinson (1967) states that Keratella feeds mainly on
rotifers such as Polyarthra. This may explain their
absence.
The next most abundant organisms were the Daphnia
and the Calanoids (see Appendix A).
Zooplankters found to be present in the waters of
Lake Mead are:
90
Phylum: ARTHROPOD
Class: CRUSTACEA
Order: CLADOCERA
Family: DAPHNIDAE
Genus: DAPHNIA
Family: BOSMINIDAE
Genus: BOSMINA
Sub-Class: COPEPODA
Order: CYCLOPOIDA
Order: CALANOIDA
Phylum: ASCHELMINTHES
Class: ROTIFERA (ROTATORIA)
Order: PLOIMA
Genus: KERATELLA
In addition, the following organisms were counted
due to their obvious high numbers in the samples.
Phylum: PROTOZOA
Order: DINOFLAGELLATA
Genus: CERATIUM
Phylum: CHRYSOPHYTA
Class: BACILLARIACEAE
Genus: ASTERIONELLA
The larva stage of development in the Copepoda is
called a nauplius. Because these nauplii are hard to
differentiate between Cyclopoida and Calanoida, they were
put into a category by themselves.
The Rotifera
The rotifers are the most important soft-bodied
invertebrates in the fresh water plankton. If we were to
designate a single major taxonomic category that is most
the class
characteristic of fresh waters, it could only be
few groups that
Rotifera. The rotifers are one of the
and it is
have unquestionably originated in fresh waters,
and
here that they have attained their greatest abundance
91
diversity. The rotifers are minute, chiefly microscopic
animals. Their most characteristic feature is the ciliated
area at or near the anterior end of the body, serving to
bring food to the mouth. The disc-like ciliated anterior
end has a resemblance to a pair of revolving wheels owing
to the synchronized beating of the cilia. We have already
briefly discussed the one rotifer present in Lake Mead,
Keratella. This rotifer is an omnivorous animal, ingesting
all organic particles of the appropriate size. They are a
limnetic or open water class occurring over a wide depth
range, even as deep as 100 or 200 meters.
Keratella generally have a spring maximum number
with another smaller maximum in the fall. Both maximums
were recorded in Lake Mead (Figure 34). The general
seasonal trend of the total plankton population (Pennak,
1946) in large, deep lakes was represented on the figure.
The third curve in Figure 34 shows the general trend of the
total plankton population in Lake Mead with the exception
of Ceratium. The fall maximum occurs when the water has
cooled sufficiently to allow the thermocline to dissipate,
the thermowhich in turn allows the nutrients trapped below
zone. If
cline to be re-introduced into the upper photic
plankton will
the weather conditions are favorable, the
and will
take advantage of the newly introduced nutrients,
ecology of any lake
show a late summer maximum. The total
is very complex, especially when discussing plankton
92
KERAT EL LA
TOTAL ZOOPLANKTON
PENNAK ( 19/46)
Figure 34. Seasonal Zooplankton in Lake Mead Compared to
Pennak (1946)
93
seasonal cycles.
It is difficult at present to determine
why the zooplankters of Lake Mead show
only one maximum.
Keratella, however, shows a typical spring
and late summer
maximum.
In the literature, most species
of Keratella are
considered rather generally as species of eutrophic waters
(Hutchinson, 1967). The numbers present at Las Vegas Wash
(LVW) (Figure 35) are extremely high when compared with the
entire Lake. In fact, the whole Basin, including Beacon
Island (BI) and the Bureau of Reclamation raft (BR) show
the effect LVW has on Boulder Basin. Figure 36 graphically
shows the difference between LVW and an unpolluted site
such as Temple Bar (TB). Here we examine the abundance of
Keratella from the surface to a depth of 35 meters, covering
over 90 per cent of the photic zone. These graphs may not
prove that Boulder Basin is polluted but they do point out
that the Basin is significantly more productive than the
rest of the reservoir.
The Crustacea
The great majority of the planktonic metazoa, both
in the sea and in fresh waters, belong to the Crustacea; in
the fresh-water environment the planktonic Crustacea are
represented mainly by species of the order Cladocera and
by species of the sub-class Copepoda.
94
3500
2800..
2100.
700
350
SC
TB
Sc
TB
BL
EB
Figure 35.
BL
EB
SOUTH COVE
TEMPLE BAR
BONELLI LANDING
ECHO BAY
MC
BI
MC
BI
LVW
BR
LWV
BR
MINER'S COVE
BEACON ISLAND
LAS VEGAS WASH
BUREAU RAFT
with January
Total Number of Keratella
and June (03Z) Values Shown
0
0
95
f:C1
04
E-1
co
tr)
tv)
ccl
4-)
ccl
o
-P
0
0
o
0
H
H
0
0
0
4
ccl
o
•
LIN
(W) Hid30
0
b0
96
All the species of Crustacea
considered here are
more nektonic than any species
of phytoplankton and are
also less likely to be at the mercy of turbulent water
movements than the planktonic rotifers.
Cladocera
The genera Bosmina and Daphnia were the only
genera found in Lake Mead. Both were monocyclic, that is,
having one population maximum and both are common open water
forms. Bosmina have their populations peak in April then
usually alternate with another species (Hutchinson, 1967).
This was found to be the case in Lake Mead (Figure 37) with
the other genus being Daphnia. Hutchinson's data showed
the Bosmina population being 16 per cent higher than the
Daphnia maximum. In Lake Mead, Bosmina were 14 per cent
higher than the Daphnia. In January, Bosmina numbered 701
while Daphnia numbered 10. In February, Bosmina again outnumbered Daphnia 4,049 to 291. In the month of June we see
that the species dominance was clearly shifted in favor of
Daphnia,
3,956 to 119 for Bosmina. It is interesting to
note here that the total numbers for these two species
greatly increase in the Boulder Basin, which includes BI,
LVW, and BR (see Appendix A).
Cop epoda
Cyclopoida and Calanoida are the two orders of the
free-living fresh water Copepoda that we were concerned
97
50 00
BOSMI NA
DAPHNIA
4 00 0
1/4
%
3000
x
1/4
\
1/4
1
\
2000
1/4
1/4
\
1
1/4
1/4
1000
1/4
1/4
1/4
1/4
1
!
M
A
M
J
J
A
--
S
-
WO.
0
TIME: MONTHS
Figure
37.
Seasonal Peaks of Daphnia and Bosmina
N
98
with. Generally, Cyclopoida are littoral benthic forms
while Calanoida are limnetic organisms. In Lake Mead the
Cyclopoid population dominate the Calanoida (Figure 38).
The Cyclopoida are generally carnivorous while Calanoida
are herbivorous. In LVW the Cyclopoida reach their highest
numbers while the Calanoida reach their lowest. This
suggests different types of environmental conditions.
Figure 39 shows the relationship of Cyclopoida to the
primary productivity data.
The total number of Cyclopoids in LVW are significantly higher than at any other location on the Lake.
Figure 40 compares the Cyclopoida at LVW with the
Cyclopoids at TB. These numbers are total counts. Copepod
nauplii in LVW may outnumber the nauplii in the other locations. In TB the total number of nauplii was 1,245 compared
to the more productive LVW at 4,016.
Ceratium
Of all the organisms investigated, the dinoflagellate Ceratium is the only one that shows a definite
decrease in numbers at LVW (Figure 41). Hutchinson (1967)
states that Ceratium is most probably characteristic of
mesotrophic water, although Hutchinson places Ceratium
under his heading of eutrophie dinoflagellate plankton.
The general summer maximum reported for Ceratium was confirmed by our data (see Appendix A).
99
SC
TB
BL
EB
MC
BI
LVW
BR
DISTANCE ACROSS LAKE MEAD
SC
TB
BL
EB
SOUTH COVE
TEMPLE BAR
BONELLI LANDING
ECHO BAY
MC
BI
LVW
BR
MINER'S COVE
BEACON ISLAND
LAS VEGAS WASH
BUREAU RAFT
Figure 38. Total Numbers of Calanoids and Cyclopoids
100
SC
TB
Sc
TB
BL
EB
BL
EB
SOUTH COVE
TEMPLE BAR
BONELLI LANDING
ECHO BAY
MC
BI
MC
BI
LVW
BR
LVW
BR
MINER'S COVE
BEACON ISLAND
LAS VEGAS WASH
BUREAU RAFT
Figure 39. January Values for PPR and Cyclopoids
101
1
(W) Hid30
102
50,000 -
30,000
20,000-
10,000
0
SC
TB
BL
SC
TB
BL
EB
EB
MC
SOUTH COVE
TEMPLE BAR
BONELLI LANDING
ECHO BAY
BI
LVW
MC
BI
LVW
BR
BR
MINER'S COVE
BEACON ISLAND
LAS VEGAS WASH
BUREAU RAFT
( I ) and
Figure 41. Total Number of Ceratium with January
June (UNA) Values Given
1 03
Asterionella
Asterionella thrives very well in polluted waters.
High numbers at South Cove tell us that Asterionella are
taking full advantage of the nutrients flowing in from the
Colorado River. Another population pulse at LVW is a
result of the enriched waters. The diatom Asterionella
usually shows a fall maxima. In Lake Mead Asterionella
showed a spring and summer maxima. Hutchinson (1967)
classifies Asterionella in his eutrophie diatom plankton
category, stating that Asterionella may be the dominant
plankter in highly productive lakes. The diatoms are one
of the most important members of the fresh-water limnetic
phytoplankton. They are always present in significant
numbers and in many'lakes they are perennial dominants.
Primary Productivity (PPR)
The rate of photosyntheses (primary productivity
example given
rate) was computed for each station. The
Island) on the
(Table 7) is for Lake Mead Station 3 (Beacon
and dark bottles were
ninth of September, 1970. The light
at 10:30 and removed at
inoculated and placed "in situ"
14 C was added. The
14:30. A total of 2.10 microcuries of
mg of carbon/
highest ppr had occurred at 3 meters (45.218
were then integrated
cubic meter/hour). The hourly data
The output was
over the solar day (686. 00 langleys).
10 k
o
Ln:
c>
^D
.•".
.-.
CD
4—
i
0
•-+
4J
•
ni"s:
1-1
I--
• J
4a
M
,
0
0
./
0
0
,
0
.-9
LJ 1,,,_.
--I
:D
CD
C)
<
M
I,
u.
1.--
0
I
.--4
I
-.0
H'
i
I
h-
,,,,CI,
4-1
c.>
0
7.:
4-
L,
C7t
:
.--n
CD
V)
1Z
4)
V,
',..
V1
..:-.7
a,
I,-
CC
0 Cq .11 C, 1) in ,0 -4
Ci ^D I , - ..C,,,,, `.... .0.0
n.... e.....
...
(.1 Ill .-n -. CD CD 0 G
ui
-,
nY
f-'
41
1.-1
X
:)
-J
V :"C
0
CC 0
V 0
.--.
7.7
VI
C".1
V) 0
V
41
IUJ
X
cP
1—,
.....{
1-.1
D
..-/
!--
In
Z
0
.-..
.-1
0
V)
CC V
41
De
0.-
0
•
1,-0
CO
(..,
7,
CD
r.., c.,
G CD
7,, , :',
0 Ca 0 0
.C - ., .•.: :, 0,.: ,...i
.--,
›.<
VI r‘l Clr,
U, ,"..,:, 0
,...) n•
.> il
CRC,, u, cr
.--; ..-.
,
Cr
NI'
0 I.- n J-, P... .-, ,..0 CV
3
r.....
3
v
...0
._., ...C. .-, ,--. V) .,..? 4-1 ,
.0
--..,
,
-..C,
n .-1 •-n -, ...,. G 0 0 0 4 in 9
0
I,.
0
11 ,-,
I.,-1:
<
0 D- 0
a ce_ .....
,.1: 0 -.3
3
3
-X
te
0
Ln
Z
..-.1
/...1
1-
ca.,-:
0
L.:
4 s. .--.
-0 ^) 9
4'
9
II
11
4
a
•
0
0
9
0
(I
3
4
9
4
4
h
9
3
4
4
3
0
...„
G G 0 C, 0 0 <D CD
CD 0 0 0 G C, 00
ro
-!, ..e,
03
D
n17.41
..4
1,5 71
.-.
FII
C-)
CD
,r1,1
r, , -zi 4, 1 --•
V
4.)
,..,!.
D
UJ
1,C
,C
_J
1•4
0
I
0
0
•-•
0
,7 .., c7 .-r re, CN.: 0 8
....,.n
.
...")
1
•
-0
3
-3
CD
CD
'•
riD
7:1
CN.I N9 7.
. Cl
fi
n-
"..
CN
P
33
n
r
.,,,T
CD
C,•-..•
j1 ,0 7, *--/ , C3 , h- C›,
4', 0
J, -13 .4.) .r2 ,-. 1...-.
C., CU ..., r'..1 ,_o 1) r,
..., ,,, ,-, Lr) rn --
DF.-+
..---..-.
c,
D
D
=
=
CL
.,
.3
9
9
3
3
0
0
9
41
3
9.
4
3
4
4
0
.s-
3
•
4
0
9
0
-.
,
1
.
9
-,
4
,-.
..,,
0
,- ,0 C,C.0 1.1,
7: --, IN ,.-, ..-- il C')1 Q C3 9 0 3
,.._, , •c, , cr f, ..,; c.,,, n. . ,-,-, ,
-I 0 , Lr, ,D .--, -...re CO CD
CSJ -, ,..1 .r) .-.:
Cr
3
f,,r- 4
9 , 0
I
t
3
.0
-3
9
3
.7%,
I- CU 1.....
O'.,.1 0
IF ,1,J , ...1
,1 -1.: ..--.
,r' ,-..,
=
In
-0
'r., :1.) .,,,".
V) 7 0
,. 0
V ,.. aV I.J
,..-,
,3
3_
"...fr:
Cr
--e
7,-
,
n:-
Ci- C'..I 41
.•-£
,T
9 1i 4, 11
-3
•
* CC 3
9 I,/ 3 .-
3 h- 0 L.1
3 4_1 9 I--0 D--'1-,..,
-D y:
3
X
D
,r ,n f, 7. ir, r... c
-.. 1.7,O
.0 Lil .,,.
o
o
c.
y a.
.
0
oCl (NI n • ---. n.I , 1" G
CV
c
C'
o
0
CU
•
3I.1-,
0 ,r," 0
4 --I. 3
9 :.-_, 0
4 ,..:, 4
0 -.1, n
V
c-4
X
7
0
=
CL
CO (2
0 0 0
I 1_,1
CD
ff,
Ln G
n/
V 0 0 Cl a 3 9
G CI 0
d: N
I.!
'D
CC 0 0 0
^
g
0 -..-. 0
IF-
105
printed in total ppr as: mg C/square meter/day = 3135.655,
or as an average: mg C/cubic meter/day = 156.783.
Many authors describe primary productivity in the
units of area or volume. Our data have been analyzed using
both techniques. The space-time distribution of ppr was
presented as an area in Figure 42.
Distributions were given for 8 stations at six
different times of the year. The ppr data were subject to
different weather conditions each day. The data were not
corrected for weather differences, i.e., cloud cover for
half a day.
We will follow the criteria for different lake types
as given by Rodhe (1969). He based his classification on an
area using the units: mg (mg C fixed/m 2 /day) (Table 8).
The January and April runs indicate the lowest ppr
values (Figure 42). However, we should note that the South
Cove (SC), Temple Bar (TB), and Bonelli Landing (BL) stations behaved as an oligotrophic lake. As we look at
classifies
Boulder Basin we see a jump in productivity which
the Basin as a natural eutrophic lake.
increases as
The June data indicate that the ppr
radiation begin
the temperature of the water and the solar
behaves as
to rise. We notice that the whole of Lake Mead
Cove and Las Vegas
a natural eutrophic lake in June. South
general, the January,
Wash have elevated ppr values, but in
o6
-,
u-
C
,'OC)
W
CJMC>
c
-MO
,- W
M>
MuIVI-J0
CM.-.
>MO
L.,
•-tAM
0-n
DOC-- COW
M",-,
0
C?
0- , -.L
UWJ00-4
'''Iûg
l
IA_ I
OD
LU
/
z,
,
v,cL
r,
1
a.
-
-
co
A
‘.0
(VJ
CO
1r
. . ,
C)
Aep/e [email protected] pox!4 uogJeD jo sweJ9
0
-J
.
1j0
OD ‘0 -7
0 0 0
107
o
o
VI
o
H
z
-H
c5
Z
o
0
Lr \
H Cq
I
I
0 NC'l
108
April, and June values are constant relative to each
station.
In September the ppr values have moved the Lake
into a polluted eutrophie state. South Cove has very high
primary productivity. The ppr drops from Temple Bar to
Bonelli Landing. However, the ppr values in Boulder Basin
are very high, with the Bureau of Reclamation Raft station
having the highest values.
By November, the thermocline is partially destroyed
radiation
and mixing has begun to take place. As the solar
begins
and the temperature of the water are reduced the ppr
at the Bureau
to lessen. Once again, we see the ppr values
Raft are the highest in the system.
to West with
The natural flow of the system is East
Overton Arm providing a flow from the North. The data on
a much
each sample date indicated that Miner's Cove had
Landing. We can
higher ppr value than Echo Bay or Bonelli
source at or below
only conclude that there is a nutrient
Echo Bay.
is suffering
There is little doubt that Lake Mead
and Las Vegas Bay.
from nutrient inflows at South Cove
CHAPTER
6
SYSTEM ANALYSIS
Multivariate Analysis
Employed in the
statistical multivariate analysis
are:
1.
Stepwise regression analysis.
2.
Principal component analysis.
Multiple regression is used in data analysis to
obtain the best fit of a set of observations of
independent
and dependent variables by an equation of the form:
y = b
o
b x
l l
9.0
b x n n
(
6.1
)
where y is the dependent variable; x1, 2' ••• are the
independent variables; and b
b 1 ,
1 , ". are the coefficients
to be determined.
A multiple regression solution gives the least
squares "best" value of these coefficients for a particular
sample of observations. The solution also gives a measure
of the reliability of each of the coefficients so that
inferences can be made regarding the parameters of the
population from which the sample of observations was taken.
For a large number of variables, any method of
regression analysis requires a large number of calculations.
109
110
Most methods of regression analysis are based on techniques
particularly adaptable to a desk calculator where a minimum
transcription of intermediate answers is desirable.
In the stepwise procedure, intermediate results,
which are not even recorded by normal calculation methods,
are used to give valuable statistical information to each
step in the calculation. These intermediate answers are
also used to control the method of calculation. Essentially, without adding greatly to the number of arithmetic
steps, a number of intermediate regression equations are
obtained, as well as the complete multiple regression equation. These equations are obtained by adding one variable
at a time and thus give the following intermediate equations:
y = b
o
+ bx
ll
y = b' o + b' 1 x 1 + b' 2 x 2
y = b" o + b" 1 X 1 + b" 2 x 2 + b" 3 x 3
•
the
The variable added is that one which makes
of fit." The coeffigreatest improvement in "goodness
when the equation is
cients represent the best values
included in the equation.
fitted by the specific variables
stepwise procedure is
An important property of the
may be indicated to
based on the facts that (1) a variable
111
be significant in any early stage and thus enter the
equation, and (2) after several other variables are added
to the regression equation, the initial variable may be
indicated to be insignificant. The insignificant variable
will be removed from the regression equation before adding
an additional variable. Therefore, only significant
variables are included in the final regression.
The inherent assumptions in the use of multiple
Shapiro (1966).
regression have been described by Hahn and
Their first assumption, says that the process error, e.1
a normally disthe regression equation to be used, is
and constant
tributed random variable with zero mean
2 for all observations. This very general
variance a e
our chemical, biological, and
assumption is also used in
hydrodynamic parameter investigation.
the statistical
The second assumption deals with
from which the regression
independence of the observations
in
This assumption is violated, yet used
is developed.
chemistry, biology, and hydromany ecosystem studies. The
complex fashion. It is our
dynamics are all related in a
hope that by treating them independently we can resolve
some of the complexities.
must also assume
In our multivariate analysis we
independent variables are known
that the values of the
that the measured values
without error. We are confident
sufficiently accurate; however, one can
in the field are
112
never be sure of the number of recording errors in a large
analysis.
scale field collection program and subsequent data
form
The fourth assumption relates that the correct
difficult assumpof the model has been chosen. This is a
to other than
tion since we have no standard to relate
in a linear
nature. We know that nature does not operate
some reality
fashion; however, we are willing to trade off
first-approximation model
in anticipation of developing a
that has in turn, utility.
the typical nature
The last assumption deals with
the data were not
of the data to be sampled. Although
was sampled "in
random samples, we feel that each basin
association of variables that
situ," resulting in a natural
that we wished to generalize.
were typical of the situation
component analysis is used to examine
The principal
structure of multivariate data and reduce
the dependence
its dimensionality (i.e., eliminate redundant parameters).
The original observation variables are
transformed into a
variables, which are linear
component
of
smaller number
functions
of the observation variables. The objective
to explain as much
is
of the variance in the original observa-
components.
a minimum number of
tions as possible with
to agitate most
Multivariate analysis tends
experience in biological
researchers who have had field
reasons are:
studies. The two main
113
1.
That a certain amount of knowledge exists about the
system and should be used.
2.
That linear relationships in biological systems are
poor.
the followHowever, all field investigators must recognize
ing problems:
1.
The inability to measure all variables.
2.
variable continuously
The inability to measure any
in time and space.
3.
in many sampling
The high level of uncertainty
techniques.
4.
of the phenomena
The lack of full understanding
governing water quality (Moore, 1972).
chemical species inter5. The state-of-the-art in
action.
in
the ignorance that exists today
If we can accept
an open mind, we can surely
hydro-biological studies with
approximation.
approach as a first
accept the multivariate
about the linear approach.
I personally have mixed emotions
Much of the
interpretation has physical meaning which is
very reassuring. However, some relationships cannot be
physically explained.
The multivariate
approach used is very encouraging;
confidence in the
however, much work is needed to build
output. The
statistical approach is used to provide some
iii
functional information in those areas where knowledge is
uncertain.
Several multivariate approaches were used to relate
the mass of data collected. Since each state had been
defined chemically, biologically and physically, the
simplest step was to linearly relate the variables. This
approach implies two assumptions:
1.
That the events are linearly related.
2.
That there is no phase lag.
Both of these assumptions are intolerable in a biological
system, but they are a natural starting point. Many
writers feel that the scope of the eutrophication problem
precludes deterministic or analytic modeling at the present
time. We agree that the level of sophistication in
deterministic eutrophication is in the infant stage. It
is therefore reasonable to presume that in the investigation of reservoirs where the time and expense required to
evaluate all the parameters for analytical models is not
the
feasible, we could try a first approach at quantifying
phenomenon. While not the ultimate answer, such models can
further studies and
provide direction and background for
predictive tools
models can simultaneously provide in turn,
needed in water quality management.
115
Stepwise Regression Analysis
A Numerical Analysis Laboratory program (University
of Arizona, Computer Center) Stepwise Regression Analysis
was used. Each of the thirty measured parameters were
assigned a variable number as shown in Table 9.
The program was set up to use the 30g parameter as
the dependent variable. This left the other 29 parameters
as independent variables explaining the value of the
dependent variable. The second run uses the 29g variable
as the dependent variable and the 30 th variable takes the
place of the 29g variable as one of the 29 independent
variables. This procedure continued until all 30 variables
had been defined in terms of the other 29 variables.
An example of the computer printout is given in
Appendix B. The equation is of the form:
y = b
o
" + b "x + b 2 "x 2 + b 3 x 3 + b 4 "x 4 + b 5 "x 5
1
1
+ b "x 6
6
where:
(
6.5
Y = PPr
b" = constant
o
b 1 ", b 2 ", etc. = coefficient
X 1, x 2' x 35 etc. = variables
Step No. 6
The dependent parameter for this run is ppr.
sixth parameter
indicates that we are looking at the
)
11 6
a)
n.r) IN- co cs• o
•
H ci 0-) 4-1 Er\
N- co crY o
H-ICICINC\lc\l(lc\INNc\In
H
H
o
1-1 CZ CZ
•ri •ri
H -H
0, 4 E
cococ)
(1)00co
ZNociaza)m
a)
I
•
I
r—I
C)
c) c)
C)
CO C7N
C
c) CD C) CD CD HHHHHH
E-4
-P
•H
-H
-P
60
c)
.H
7:5
u)
o H
cr3
o
) z
Q)
4:2
czs
•
(16
•
•H 0 0
H H
O 0 4-> ,O -P
0
Sai tn
E
bl)
CD Cn'l
Q)-1-1 .H OHOOUOH t-Og g0
a)ozcnoZ
0 Z 0-1
117
recognized as being a major factor in ppr. Each parameter
1, 8, 9, 14, 23, 25 has been independently investigated by
the computer and by a stepwise fashion has moved to the
next most significant independent parameter contributing
to ppr. The program was set up to go through the 29
independent variables or steps printing out the decreasing
contribution of each of the steps.
The F level in Appendix B was 4.1. At this step
(No. 6) we choose to analyze the output. The R-squared
value 82.2 per cent showed that we had explained the
variance in ppr very well at the 95 per cent confidence
level for the parameters used.
The six variables were:
1
Temperature
8
HCO
9
SO4
14
23
PO4
25
3
Bosmina
Cyclopod
step No. 6
Using the constant and coefficient given in
ppr was:
the equation relating the variables to
ppr = 117 + (.527)(Temp) - (.593)(HCO3) - (.090)(504)
+ (235)(PO4) - (.035)(Bosmina) - (.014)(Cyclopod)
(6.6
)
118
The equation for ppr shows that the independent
variables are a function of one constant plus or minus six
quantities. It is curious to note that the computer
assigned priority values to a physical parameter (temp)
then chemical parameters and finally grazing zooplankton.
The positive effect of temperature has been widely
confirmed. As the temperature increases, the ppr will
increase. However, the HCO 3 has a negative effect. The
inhibitory effect of SO 4 is very curious since SO 4
increases considerably across the system. The strong
argument about PO 4 as a controlling nutrient gains some
support from the equation. The grazing effect of Bosmina
may reduce the phytoplankton present resulting in the lower
ppr. Cyclopods which may eat plant or animal material,
negatively act on the ppr.
To confirm that the equation for ppr does exist for
the available data we chose the results at Temple Bar on
September 10, 1970. The parameter values at that time
were:
Temperature
26.6°C
HCO
142 ppm
SO
3
4
300 ppm
PO 4
.01 ppm
Bosmina
0
Cyclopod
1
1 19
applying equation
(6.6):
ppr = 117 + (.527)(26.6) - (.593)(142) - (.090)(300)
+ (235)(.01) - (.035)(0) - (.014)(1)
= 22.2 mg C fixed/m 3 /day
(6.7)
The actual measured ppr on that date, at that
location with those variables was 18.7 mg C fixed/m 3 /day.
These results confirm the excellent predictive ability of
the equation for the given data.
To prove that the equation held not only in the
summer but also in the winter months, a January run was
analyzed. The Beacon Island sample on the 27t1 of January,
1971, was recorded as having:
Temperature 11.6
HCO 156
So 4
3 20
3
02
PO .
4
Bosmina0
Cyclopods 0
The equation for ppr was:
PPr = 117 + (.527)(11. 6 ) - (.593)(156) - (.090)(320)
+ (235)(.02) - (.035)(0) - (.014)(0)
= 6.3 mg C fixed/m3/day
(6.8)
on the 27th of January
The actual ppr value recorded
C fixed/m3/day.
at Beacon Island was 6.109 mg
120
The examples indicated that the equation held for
different stations at different times of the year with
different parameter values.
Principal Component Analysis
Since one of our objectives was to functionally
relate all of our variables we decided that a Principal
Component Analysis would tell us which variables were
accounting for the biological response. Only the
5111
and
30 meter data were analyzed because complete chemistry
studies were done at these levels.
The computer program referenced the 31 variables
(including depth) and normalized the data to a zero mean
and unit variance (Table 10). A large 31 x 31 correlation
matrix was set up (Table 11). From the correlation matrix
solution the eigenvalues were found and their per cent
contribution was recorded (Table 12). The eigenvector
showing the best correlation accounted for only 20 per cent
of the variance on the data. The plot of the first
eigenvector was given in Figure 43.
Generally the plot indicates that at larger depths,
extinction were
both the dissolved oxygen and the light
and electrical
low. Under these conditions soluble salts
was supported by Na,
conductivity were below normal. This
response to this
Cl, and SO 4 being low. The zooplankton
121
Table 10. Principal Component Analysis -- Lake Mead Data-5 and 30 Meters (31 Variables).
Index
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Mean
Standard
Deviation
17.50000
14.38043
8.94674
12.50000
4.76099
2.36517
12.92717
768.18478
1.12402
17.64602
1.23913
3.45543
6.03499
1.99931
153.63043
301.06522
13.09391
60.59036
98.18478
32.02174
80.85870
103.56522
0.01098
4.93185
0.08543
0.04043
0.03913
0.07620
0.41250
38.32609
20.93478
21.69565
9.06522
32.68478
20.90217
9.77174
563.91304
4.57282
13.41196
6.57884
5.33442
9.88035
7.32670
0.01153
3.95796
0.04340
0.03025
0.02648
0.06437
0.22305
63.16922
55.66804
44.53473
18.63651
80.613 00
35.88389
29.82015
1619.59859
8.77 0 74
16.20294
39.63884
0.06873
Variable
Depth
Temperature
Dissolved Oxygen
Light Extinction
Soluble Salts
Electrical Conductivity
CO 3
NO
HC 3
SO4
Cl
Mg
Ca
Na
PO4
pH
Fe
Mn
Zn
Cu
Nauplii
Daphnia
Bosmina
Calanoid
Cyclopod
Keratella
Asterionella
Ceratium
ppr
Alkalinity
(.4.
122
4
.
r--
0.1 45, 441 P, tr. (4
r
r-
(I.
r,-
•
op
•
•
p•
•
I
N. It Fr,IC C. C. Ir
-1 If C' Un
o •:- Ct N
4,
r.r
.11.•••
••
.19
•
•••
ag•
III!
tor
•••••••••0
I
•
O
r'il
C
ORO Li
C
CV INClr'fr.0 C'Clr
C
•
•
.(.• C
4111111
C
f N Ct. (7' 4f..
,-
1:
I
.111
4474
lr C
C
J. C. C. , (N.:
C
C
Cl •- n
• c . . .
i
11
• •
c. •-•
7- C.7
c;
14,
jr. 4. 1 it. r, N. fi C., O..
C., Cr
. • •
• • • • •
1•I
.-
J.. Cr
n • • • .
. • •
.
.
•:
.
.
I
.
'
. • •
1,11
1111
,
n
1
111111111
,
I
c
C
C
, . •
1
..- i
r.'"
..•••
111
11
41
, it C-c
•
• . •
1
O. 1'
11
cr.•
h
tc",
Ili51.-1/1
C'
if UI CC ••••4 C
n••••••••••••••.....
•
(c. r4 .-- ...-1 (N 4.4. I' C N -, Cc 7 .., -.1 , LC. C' n... Cr (' C r-. h CI fc .-- Lr -- --.., r: -- n r. r- "- 1.- ,e. lv I. ,I , ir ‘-' C7 ..— f
, . il A. -- ,..' c 4- ..-. - c.- - r- 4 :
r
••
I
.
;-
II
;
'I
I
i
.
I
'
.
,,, ,.., ..-. ,N, r- ,--
1
I
.
..
,.
'
.
,
4--r,
0- -3 - 0., ..,. r* !-- u-
t'--,
f, .4
-'C -e
. 4
•
a
—
•
• . .
. • •
1
. -. Lf
—
CP. U. '
,
'.
,,,
11
-.1.
C'C
C. C
IN
.-
... .....
(.,
1
1
!
.
.
.
4
4
11
14
- .7
.
Cr'c C
/1
(..
.-
. • • .
r
/
N
•
11
N. h C,
•
e
•
.
‘4.-..4-
•
tf,
.
.
-
.
1.
1111
,
•
,.
•
:
,
!,
I
1-- (I
,--, r .
'
r. , ,, 6 7 (,... u --
-,1
1114
1
'1
(-•
.
.
.
(4. , r, zi , c.
.
I
1
11
.
.
..- , . i- (- C If C‘. .f N , .- If 0 . ,_, r. c -, .4. C l7 ,
C n c 4...- :N. c. Ly . ,.- . .- (... „I- .C.. CN ("e, •4. - c ,, r.- rs - c, 7- ,
!
.
c:
n1•4•••••99•0,11••••••••
ta•il•••
a
'
'
Ll-Cl./1
II
1
1
IIII1
1111
•
• 1
(N.
-
••
9499,0
•.••••0•9•••10s
•69
•
-a
II/
1
•
r
F
(
•
I
tl
r
(fit
•••
111;11
C
12 3
I
40
•
1
I
•
4077
1
1
.- ,
•
n
r •- In 1 -•
r n
n
•
•O
•••.••
•
I
-111111111
I
•
C
tl* r u o < Cr ci if nr
60
•
Cr
cf. c R C C! I( 4 11 lf • c_-• Ç c' C c
• -
n
••••
••
Ade
.a
.0
•
•
•
.113•••••••••
OU
1
11
111111
t
g
g
,
CO -1 . -1- ..-, .d. 1 , c c- c 1 Cf C -,
CV If n• rn n• cr .-., cc n n •C 4 c• .n. n. cr. cc r.--
C:: 7
.
.. •
•
•
.
111
I
• •
s
I
11.
•
,
.c - n" 1.,
c C
c
• 1- , 1.;
• .
.
•
1
,
1
..
•
]
'
.
.
...; I. -
c. 1-
r-
• .
.
• .
111111111
1-1
111
•
•
• .
. .
• •
.
.
• .
I
c -• -; c rn, r C. - r .- r rn -1 if:
c: c c C. c .-- C., -- ---• n c. c c. c - r.. 47 • 0 •
7
• •
7
• V
•
•
•
• 7 •
4
1 -11,1i,
- ;
1. 1 1
;
I
;
!
.
• .. .
•
. • •
•
. . •
. e
,
111111
i
1
i
,
•-•' a
,
1
•
;
•,
•.
.
- r .,r.
Lc c. ...r. C\ C- - ..r V. C .
-; c• c'• , If'
,:
..; cl . n. 1.r.c: in r • c C n.
, n 4' r-- c .--• .;" r- •-,:- .. ..., r. in tr .... c z/ L.T c - C\•^
cr. r. n. c ..7. r.. .-n.
•-•
c...
C.
r..
.
.
c•
n.
c
c
n
c
e
••
0
c• - N. c. rr n C- C. c :-. .-- c.
4103.
. . .•
, r': r ,
- -
.
. •
• • ; .. . .
.
.
I
g
4
I
I
0 1, ., c.
.
. .
o, c• 7,
4 . •
;
1
1
T) -.
I
I
I•gl
-
•
1.-- ' r- 1,- .--• c
1.f'. ...c.,
a
c. , c 0 ..--, c ...," c-- • c c.„ .
I
1 .
•
,
1
I
•.
.
.
•-• 1
I
g
•
,
.
.
t
._
I
I
1. r CC
c 1
n
• I
I •
.•
•
,.
.
•
•
.
•
.
.
,
•
.
.
•
•
,I
g
.
'
.
..
n
•
•
a
.11.01.
I
II
•g
-.4.
•
•••••••••
I
g
I
.
n
11
.
.
.
n
.
.
.
,
.
1 ;- -- r.- , .7
.1- ...1-• C' C\ ... 1 Ci ..,1: --- 0 n• -- 0 n. C
.4...;•••.•••• •—I
I
I .,II g'
.
nr- r•-• rr •r- 0- n. c -- n• c n c r; ..- , ..” n c
••
•••10-13.00 g
.
,
.
tr. 0 -.' .1 r - ,---• - r..
I:
I
I
g
.
.
.
,
.
•,
,
.
g•'-'.1',..
0—
'
V ,4-
3••••••
ee
•••.1•0••a••• .
g
.
.
n
g
g\.,gl.
-
.
I•
.
,--- c... c rat -.! ...c• -. 0. Lr Cr
- --. 0 ;Sr X"<
CY' ON -• f-- 4 N -- 1- •- -1 n..
'I
,.
0. - _ ,
c c,
.
'.
• . •••••••••,..
Ill
R.
•
,
--C. -,-4.
;,-. ;'... ‘-_-
-
-
I
I,
r.- r- ‘f
0 ,. ,/, c. r. c. CT ..,- L ir.- rr c ;- '1' I-- CA
c , C ...' If tr .... C.! -, ..,-; c .- c r - c ...I1- , c •
•C • r.:' -, (--- is' “ , ..r. : ,
C
Q
'..---.
all
•
•
0•
4,00'0
•
I•
C c -r
-;!
r\
A •
c(.TCrit C. -t'
.01
7—
nn ..;* a:
1--- C
D
L,
•01.4,.;n
•
I
n
,
,r tc ..1-
03mIc
I
A
•••
a
gTI
I
.
i
,
.2
if n
n.-
c
nn. r•-•
7I
;
I
!
124
•,
'
I -IlI , I
,
1
1
.
I
.1111214
I
I
.,
•
...1 V.' c_. c'•
-.I c -_ a
0: 'III c 4r r: .--, IN •-• fT, C C CC N CI' .I' It- .--, C 0 C L.4 0: h. ‘.0
I
r- I
I
. •
•
.
1
1
.
. .•
i
1
1
'
.
I
I
.:r1./
4.,.
,
I
I
u , ,.-.• r
4.. ,,f• c
•
.
g
i
,- rT 4.-T •4. 4-- . - ...•
.
•
I
i
.
I
'
i
I
.
. a
.
4 .
.
•
.
.
.
• •
..
•
I
a
,
.
I
.
a
I
r
,
,
•
a
...-
,
II
II
.
I
I
•
1
•
• .
• . . • . .
•-•
I
. • . .
1
. . .
I
• . •
1
. • . • • . . •! . .
III
1,
!
i
;
I
,
•C !>-
!
. ; . . -:
'
a CI
i - II " T I' .2- r- C• 7- c - I-..
.
.
.. 4- /4; -- If - .: c - c.
a' IT 41- r - r...: ...1 c It f,
if -J4 0 t 1.1 - (I 4-
0. Lf. 1, , -- c - 0 ..- 2. -- 4... 0
.
;
.
1
•
I
ln
1
i
i
r
0
I.
;
• : ;
1
1
I
r- IV
Nr
:
• ..
•
., .4.
2,2III
•
.
I
; ;
;
.
- , h .e. ..0 , .-- c" 0 st If.
if if 2.1- IA I- - c h ,T Cf sf N , .2
r- C. •- h 4,.. ,-- 4.- 0. -..1'
..-. 4.f ....- ... 4- II-C 0. IN l• --., 4
.r.- c. -- r - c -- c.
• . •
•
I • ; ; •
1
•
0-
C
c
c.
.4'
r 4.• h •
a
1 , t
1
2
.
Il
C • IT- ./
r:
I' 4 sf- C
n,•••••
•••
00•4
•••
4.0
• H••6•04.-•••••
I
1
1
I
r-'. ,Ç IT C A .. ...-. f - 4_ 4 c 4
a .- 0- kr ir h 4 - Tr. ..( r4-- C
T. c .- h 4, 1- 4';
If -4 --. h V ...-• 0
LI 4' t" . - r N r ,
, T. h 0 if IT c I. ...-4
4., a
CT....4 Er c iir II . , 4 f
O. fn ..C_ 2C r, 0 a , - 0.
-_ -- .-- .- 0 r -- .-"
t
0
0..
..c
4
.-•
ca
0.
4,..
..
c_ , CD
•
,
I
ii
I
,
.
1
t
'
IT
I
I
1
I
1
1
1
I
1
I
1
,
.
I
n
1
!
.
I
,
:
- (..- a c fr c.., ..,, c -i C.: .t ,,
...7.2 C ...( ..- t. t - C C .- N ....., ,,,r
rc
(T, IT- ..27 0 CT (^ - --. • n-• IT C.
, 1- • - , ,- l.,- , C ,-. r
0. .4 4 C I.- C • C C. C 1 - C. , 0"
, ?
oa.'•••••••7 •
I.
,::
.
',.
i
!
?
I
.
•
n•
'•
i
I
..
1
i
a a'
0. ::-.
C", •-• •,... 7- ..r. 0.. •- -t
I
•
Il
.
r..........•
1
:
•
I ,
c-- ..a
0.. r_,
•
•
•
•
I
•
•
{
I
2
. --
1
I
-
h r_ .L. C
I
4
I
.
;
,
1
,
•
I
Cs f h h 2*- - 7. c. ,i 2-, !
', ..1. •-• f• - a: 4 L. (.• 4./
•.1-
I
I ,
I
.
,I•
..
:
,
a
441
GI•
.
I
1
4.-- h c 0 c ,. ..2 -0 -I. N 0.
.4 , IS .4
I: tr. C' , :x ..! .4' ....-2 4
- a - ":: •- • I.! , (,. al:
' ..-4. h h h- N IT. c ,
, , 4-• r ' '.J ,,
4
-,
..- , ,
.
.1
_
I
0.; Ai 1.IT. .- -. in C,
•
1111
I
I
(T
I
/
•
I
'.
....6
, I
•
.
..
•
.
,
r
.
.'
a .
.
•
n
•
,^ Lr ., c.- -
.
a
.
I
a.
•
•
•
. i
•
I
'
.
.
...4- ...:2 0 r
'
!
.
•.
•
I
:
, •-
,
.
.
..-- .. n•
I
I
.
.
I
•`,..,
1
I
I
I
I
I
I
11111
.
:
t
i
.1
I
1
i
:
.
1 25
Table 12. Per Cent of Variance Explained by Each Eigenvalue
Eigenvalues
Per Cent Variance
Cumulative Per Cent
1
6.00110
.19358
.19358
2
4.91945
.14579
.33937
3
3.9484 0
.12737
.46674
4
2.52296
.08139
.54813
5
2. 0 8846
. 0 6737
.6155o
6
1.84646
. 0 5956
.675o6
7
1.25015
.04033
.71539
8
1.21887
.03932
.7547o
126
L.)
127
deep water was a considerable reduction in the numbers of
all the species.
The second eigenvector accounts for about 15 per
cent of the variance in the data. The plot in Figure 44
shows that close to the surface of the water the temperature was above normal. The light extinction was very high.
The zooplankton were above normal in numbers with the ppr
being very high.
As we looked at the next 3 eigenvectors we
realized that the data were separating out according to
depth, season, and location.
Since the best eigenvector could only explain 20
per cent of the data relationships we decided the Principal
Component Analysis was a poor predictive tool when dealing
with seasonal data from different stations.
Testing the Model_
Since the multiple regression model accounted for
82 per cent of the variance in the ppr values, we decided
Reclamation
to test the six variables used. The Bureau of
Raft on the 7t.J) of September, 1970 was eliminat
,-
d from the
data. The Stepwise Regression was forced to only at
resulting equathe six variables of equation (6.6). The
tion was of the form:
128
,0
S-1
c6
CD
o
S-1
o
CD
b0
0
•
—14
bp
129
PPr = 116.5 + (.529)(Temp) - (.589)(HCO 3 ) - (.090)(SO4)
+ 234 (PO4) - (.035)(Bosmina) - (.014)(Cyclopoda)
(6.9)
The variables measured on the 7g3 of September, 1970 at the
Raft included:
Temperature
26.40c
HCO
117 ppm
SO
PO
3
4
320 ppm
4
.02 ppm
Bosmina0
Cyclopoda 0
Substituting into equation (6.9):
PPr = 116.5 + (.529)(26.4) - (.589)(117) - (.090)(320)
+ (254)(.02) - (.035)(0) - (.014)(0)
= 41.0 mg C fixed/m 3 /day
(6.10)
The actual ppr measured at the Raft was 37.8 mg C fixed/
mg 3 /day. Since equation (6.9) was derived independent of
the data measured on the 70 of September, 1970 we are
optimistic about the excellent predictability of the
regression equation.
CHAPTER
7
CONCLUSIONS
Discussion
The ex post appreciation of the sample design in
any field study should be evaluated not only to qualify the
ex ante design but also to provide desperately needed
insight into future field surveys. The Lake Mead investigation has shown that repetitive winter investigations do
not provide further insight into the ppr causal relationships. We conclude from the chemical and biological data
that the February run was of no added value. Rather, we
suggest that more sampling should have been done during
times of greatest change, i.e., August to November. The
temporal sampling design in any ppr investigation should
be based, ex ante, upon the knowledge of thermal stratification in the subject water system so as to insure the
sampling of highest pollution potential.
The horizontal spatial distribution of ppr did not
led us
vary in wave number with temporal sampling. This
to believe that we had adequately described the spatial
changes in ppr across the reservoir. We were very interstation did not
ested to note that the relative ppr at each
if the inputs to the
change with time. This suggests that
1 30
131
system do not change, the water temperature is a dominant
factor in ppr.
The vertical distribution of ppr at each location
was described very well. The intensive sampling at depths
of 0,
1, 3, 5, 7, and 10 meters quantified the areas of
highest algal photosynthesis. The samples taken at 15 and
20 meters described the reduction of ppr with depth. The
ppr below 20 meters was negligible at all sample dates and
locations.
The assumption of chemical homogeneity above and
below the thermocline is too gross an approximation for
pollution investigations. The cost of chemical analyses
was such that samples could be taken at 5 and 30 meter
depths. We would rather have seen more chemical sampling
in the high ppr areas. Future investigations should include
in-depth sampling to develop some appreciation of reservoir
effects upon chemical distributions.
Temporal and spatial changes in chemical and
biological properties of the river-reservoir system have
been established. The ppr data in Figure 45 show how Lake
Mead behaves as an oligotrophic lake in the winter months
and as an eutrophic lake in the summer months. The sensitive ppr technique indicated how the algal growth rates
accelerated from June to September. The EPA (1967) Report
in May, 1966.
on Pollution in Las Vegas Wash was conducted
EPA was attempting
Using the poor index of 2,000 algae/ml,
3.0
0.0
Location:
Beacon I s 1 and
Figure 45. Temporal Changes in Algal Growth Rat es
133
to set some water quality standards. Our data show how the
September values are considerably higher than the May-June
values. Consequently, any standard set for May values
would naturally be broken in September. The Bureau of
Reclamation (1970) report used data taken in May and
November, completely missing the high September values.
The Bureau of Reclamation (1971) report used March, May,
August, and November sampling dates but showed that highest
chlorophyll values occurred in May, not August. The May
temperature profile showed the Lake was still in winter
conditions with no thermal stratification. The EPA (1968a)
report used data taken in April. We can only conclude that
when pollution parameters are being investigated they should
at least be measured during their highest temporal values.
The spatial changes in chemical and biological
insight to
properties of the system provided considerable
the effect of inputs to Lake Mead. Las Vegas Wash, the
all conMuddy and Virgin Rivers, and the Colorado River
chemical species. The
tribute different concentrations of
effect of this changing chemistry could only be measured
sites. The input and output
by using a number of sampling
often acted very much alike.
(South Cove and Bureau Raft)
system described variaHowever, the changes through the
problem in Lake
bilities that partially explained the
Mead (Figure 42).
134
Primary productivity rates (ppr) and plankton
population counts were conducted at several sampling locations in Lake Mead. Two major sources of pollution were
concluded from the ppr data (Figure
46). Las Vegas Wash
was the major source of pollution to Lake Mead. The three
locations in Boulder Basin indicated that the pollution
source at Las Vegas Wash had affected the whole Basin. In
fact, the Bureau Raft, which is the station closest to
Hoover Dam, shows ppr rates higher than Las Vegas Bay. The
higher values at Beacon Island which is upstream from Las
Vegas Bay, indicates that the hydrodynamics are circulating
the pollutants throughout the Basin.
The minor source of pollution occurs at South
Cove. Since there are no sewage treatment facilities at
South Cove, we could assume that there is some contribution
from "campers." However, earlier works by the EPA (1968a)
indicate that the highest nutrient levels are recorded
above Gregg's Basin. We conclude that there is a nutrient
source in the Grand Canyon. The source may be natural or
the result of human wastes. Regardless of the source, the
need exists to determine where and how nutrients are
entering South Cove.
The zooplankton analysis at each of the stations
has resulted in useful diagnostic data. The total number
indicates that Boulder
of the rotifer Keratella (Figure 35)
of Lake Mead
Basin acts completely independent of the whole
135
3. 0
.03
.02
.01
.00
1.0
BR
LVWBI
EB
lic
BL
TB
Sc
Fi g u re 46. Spatial Distribution of PPR and Phosphate
(September, 1970)
136
with respect to Keratella. South Cove and Echo Bay have
high ppr; however, the total number of Keratella at these
two locations is not different from the low ppr areas at
Bonelli Landing and Temple Bar. Keratella is an indicator
of the poor water quality in Boulder Basin. The increase
in this organism outside of Boulder Basin would indicate a
spreading of the water problem.
Ceratium was one of the phytoplankton that were
analyzed. Hutchinson (1967) places Ceratium in his
eutrophie plankton category but states that Ceratium is
most probably an indicator of mesotrophic water, i.e.,
between oligotrophic and eutrophic types. Our phytoplankton
data have
data confirm Hutchinson's theory. All of our
and the rest
shown that Boulder Basin is highly polluted
The total number
of the Lake is in an oligotrophic state.
in
of Ceratium (Figure 41) shows that Ceratium is low
in
numbers in Boulder Basin indicating that it can exist
of the Ceratium
eutrophic waters. However, the majority
was found upstream of Boulder Basin in the oligotrophic or
mes otrophic waters.
concentrations at each
The macro and micro nutrient
The concentration of
sampling location were determined.
the ppr rates.
micro-nutrients was not low enough to reduce
SO4, and PO4 levels
the HCO
There is strong evidence that
3'
This assumes
the ppr.
are the major elements controlling
1 37
that the other nutrients, i.e., nitrates, are available in
sufficient supply.
The two dimensional distribution of the nutrients
has shown that Las Vegas Wash, Echo Bay, and South Cove are
receiving high nutrient levels. The vertical distribution
of nutrient shows that Mg, K, Mn, Z, Cu, and SO 4 do not
change with thermal stratification, while Cl, Na, Ca, and
HCO
3
show a noticeable difference in their
5 and 30 meter
data during summer stratification.
The last objective of the study was to relate ppr
to physical and chemical changes in the system. The
sampling dates (Table
6) were chosen to maximize the number
of state sets. The surveys were designed to measure the
state sets as thermal stratification began and ended.
State sets were also measured during stratification and
winter turnover. The locations were chosen across the
Lake to represent the largest variety of conditions. In
effect, we tried to vary all the parameters over their
natural range in Lake Mead and relate them to ppr. The
four classes of state sets are given in Figure 17. Class I
included the highest number of variables and was used to
develop our regression model. All of the variables included
of
in this study are represented in Class I. The number
Class I,
state sets in Class II is four times as many as
Classes III and
but only 14 variables were investigated.
138
IV have progressively higher numbers of
state sets, but the
number of variables are reduced to two.
The Class I data were used to develop the stepwise
regression analysis. The resulting equation was:
ppr = 117 + (.527)(Temp) - (.593)(HCO 3 ) - (.090)(SO 4 )
+ (235)(1'0 4 ) - (.035)(Bosmina) - (.014)(Cyclopoda)
(7.1)
The computer selected physical, chemical, and biological
parameters in their respective order. The units to be used
in equation (7.1) are:
ppr
= Mg C fixed/m 2 /day
Temperature = °C
HCO =
3
ppm
or mg/liter
SO
4
= ppm or mg/liter
PO
4
= ppm or mg/liter
Bosmina=
number in
6 liters
Cyclopoda=
number in
6 liters
The equation accounts for 82 per cent of the variability in
the ppr data. The level of confidence in these six
parameters was set at
95 per cent. The works of Cole
(1967), Chen (1970), and DiToro et al. (1970) describe
algal growth rate (ppr) as a function of temperature,
chemistries, and grazing zooplankton. The major difference
We
in our approach is the lack of a solar energy source.
139
have shown that in Lake Mead light penetration improvement
has not resulted in higher productivity. In fact, as the
light penetration improves from South Cove to Honelli
Landing, the ppr rates decrease considerably.
The equation (7.1) was tested by removing one
station from the data and computing a new stepwise regression equation (6.9). The new equation was very close to
the original equation which included all of the data. This
would indicate that the pattern across Lake Mead had been
very well defined. The result of the test was very close
to the actual ppr value. We conclude from the test that
the state sets measured included the ranges of the variables
in the test station. We are confident that we have measured
the majority of the state sets in Lake Mead and can predict
the ppr at most locations at any time of the year. Using
regresthe criteria established by Rodhe in Table 8, the
realistic predicsion model is a first approximation for
tion and is of management utility.
Future State Set Analysis
holds for the
We have shown that equation (7.1)
meter depths. However,
eight stations in Lake Mead at 5
equation can be improved considerably by
the utility of the
Class I. To
increasing the number of state sets
we q1d maintain the
increase the number of state sets
number of
same number of variables and increase
,
140
different locations. To accomplish this, we could take the
study into Lake Mohave, Lake Havasu, and Imperial Reservoir.
At each of these locations several sets of states will be
obtained.
The validity of the existing model should be tested
with a complete set of independent data. The test data
will serve not only to validate the model, but also to
increase its predictive range.
APPENDIX A
ZOOPLANKTON DISTRIBUTION
1 41
0
•
142
0,0 C g
O U) 0. 0 C 0
'7 LL, .‹ '7 Ul 4
N-
7. -0
N In O :A
CO c",
C)
\12,
CO
CV H N-
0'O
g
O 0
CY, 0
,0 for,
Cl 0
01 C1
•
N0'-0
0
CC
O
u)
C' U.) "..?
cf)
<
Ll CI Ng ON
Cl
H
•
CC'
ON -7' 00
C.1
Hi
k...D CI",Cl
c"-N
H
CN
N0 CI
COO
C", 0' 0 CO
H CI H
H CI
-
Cl C 0
0' LC' ]'-
N zr
tH\
Cl
C's
CI 0s -0
N N
C.0
C",
H Cs 01
CI H
N. UN 0 0 H
c•-n
v-1
c"-s. c•-n
n—t
O 0,.0 H 0 0
c•-n
c•-n
c•-n
cV
• C
H -7 CC Cl
O
c",
C']
N- ir \ O 0
O H
CI CO H
•
• 2
00.000
HI H H
0 ON N- H 0 0
-7 0 CI
O
O
•
CC'
CO CO CCO CA ON
H CI "-0
.
L"n H Ci
•
0.0 L., C.! oC.
n
OCOCCO H 0 0
Ul 0 0
-7 CI
C
CI IN- CA
bl)
g
g
r, 0 0000
cl)
--, -^ 7^
LC'-CC' In
0' 0'
H
7,
c,.
CO H H c., 0 H
H
In N 0 H
Cl
O H
‘,.. N.0 CO
H
0 CO --".
L'l N
,..")
ND
COCO N0 c0
ON CI 0
Cr,
N 0
CI ("1 H
N-
0
0
ill U-N 0 0
in N.
Ul
C1
Cl LI- \ C., 0
tH. 0' -7 ,-0
IC'CONI"-
1-1 H
000
CC' CO
-Z'
Ul
CO --.. -0 ,.0
El 0
H
O
(N- CI El
N
E
U)
O
CU
.74
O
0
.,-;
b.°
C-
O
H
0.
g
0
7:,
CC
0
HI
g
4
,0-.
0
0
CC
g
HI
E---
Il)
0
r0
0
HI
CC
N Lrn
0
143
r. -0
C
C
C. 0 C. C., 0 0
o
=
C I 1.31 _7. 71 0 -7
-t" -7 131 1 . -- l'.- Cl
HO
C 1 H
.0 ',...
00
-.
0: 0 0- 7. 0 0
0300-010
‘4)
cr)
cl 2.
C
c
0. 05
n
1-1- 0 CO 3:7' -..2 C
0 ( 1.- 3.7"'....0 0 0
C111- 73 71 H
7.,1 -.7 CI
H
ClCO-C'
,-I01
0
3/5
-7 CO 71,0 '...0 CO
-7 0 C1-7
I--4 73 0 I' -01
CO
Q
H
r.
C
3f)
.'2. '..n
cz
OC
H
..-i
I.:\ VZ CO '...5 C CO
CO n CO c 0
H '-0. 0 C-, COCl Cl -z.
Cl
Cl Cl
H
Ht
0 CO oi , ....2 C 0
H N-
CO
COH
Cl CO CO H 0 0
--." f-, r- ci
cc
Cl
br.
C
--J -I
....1
31 05
• C.'
0 CO %.2. ',..2 , ..2 CO 0:,'21.7.2 1- CI H.
0 , ...2 CO -.7 0 c
Cc c •_-;
,--i
r- u-‘,--! ,--, -7 COCl
CO COH
COc-,-,
O C
22 ,_
cn
\45 Lrl C l'- -CO COC'. CO Cl 71-7 C
COCO 1:-.,. — --..-7.
,...0 CO t.-•%-.7... H
O
0 -td -7.' cc -.7. 6
('-.. S. 171-7
H
71 71 0 ZO
'CO
-7' Ir- C I
4 ./.
O 0H
n..3 .--0-•
C C 01 , -2 0 C
CO
y-
CO
r- , ...2 CO 0.1 cr,
H N- CC CO
Ol 131 Cl 71
CI
-7
71
01 01
Ln
I''.-- H n 71 C.-. 71
71 L'1.7. , .7". 7I .7.0
C C H -7 C C.
-
1.--1 5..-- ',C. C..5 70 I -
CO CO371 L", C I.-
H
-7 n CO.....7'
71 CI 1 - 171 r- H
1., 0
.--: , ..f. l ".- .O,
C: ',..•01-C' CO,.-COCl
•
0COCO
Cl
0
C 7.3
0H 1'1- C 1-51 -7 H
(3 :3
Cl ,..: CO 0 -7 -7
O H
COI--
O
C
11- 'C. 71 0: 71 7.0
C1 L".. 71 -7' -7 CO-7
CO
n.:: ...2
0 H
f"--
HH
O CO - CO Cl CD
CO
CO
co
if:
= 1-i
1-1
C
N
N -,..- 0 H,
--- - i '''. -
2,
1,- N- CO ..2, -.7
0 ...
•
2,CI Cl c.: C
1 --
CI COCO0101 h- Cs
H
H H
:h ....?:
0
-4
• 035
7 71 C 07 71
7. 4, L^, ....
0. L-
H CO r- ..-.- -7 ...:
• C-0.0
0 iC0
-
,...7 0 -7 L.-, 0: --.CO
CO
71 1 1- -7. 1 -- N-
C.7" 1.7",
c n :•-n C
H
c n H C C
C
0 010CC C
H-7'
c
L-1 71 C
Cl CO
-0-
073
O
APPENDIX B
STEPWISE REGRESSION PRINTOUT
1 44
145
O
O
(1)
Z
r-I
(5
ON
El
0 0 0 0 0
0 0 0 0 0
O + + + + i4.1 W C.4 W W (4
.O
.-t .C:)
0 \ ci r- cr'N 44-
Cq H
000
+ +
(4 (4
0
Ce\ N
C*-1 Ii"\
OE'
en 0
u)
f'-- H
---t
g
0 C'sl
C13
• •
(1)
t'-- c\I
,
-
r-I If \ LO C\1 c0 C-•
en --I-, (-,-) Crn r-- --ti
-t Irn H en --.1-, en
LI \ --1-4 ir \ N 0
00 0-n ,--1 ,--1 0 l'--
• • • • •
•
N n..O Ce\ Ce\ N r--1
I
I
I
I
C\1
Cl
0 0 0 0 0
I
I + I
I
I
(4 (4 (4 (4 (4 (4 O
co LA cv )
-
0
Cl)
0
0-\H
cn
c•-n CrN r--I 0••
(5
cl
c\1
cy-N
0
(-q
(Z
C71
If\ 0.•
CO C,-) CO
•
• • • • •
H CO
N
CH
0 4-1
O
ri
5
En
o
r-IHNNNN
O 0 0 0 0 0
I
1+1
1
I
T-1
CZ
*4
w (4 (4
Cl (-r•\
k.0 0'.0 N c0
c--\ 0
0', CO cn
k,C)
c'--sCV
C')(---)
•
•
•
•
•
•
IN c3n
I
I
I
co oN—t n
,
N N
n.0 Or\ 1-r
0'1 •-t4
REFERENCES
Standard
American Public Health Association, 1971.
and Waste
Water
of
Methods for the Examination
Association,
New
Water. American Public Health
York.
Evaporation Theory and
Anderson, E. R., 1950. A Review of
Water
Development of Instrumentation, Lake Mead
Navy
Loss Investigation, Interior Report,
Electronics Laboratory Report 159.
1951. Physical
Anderson, E. R., and D. W. Pritchard, Navy Electronics
Limnology of Lake Mead, U. S.
Laboratory Report 258.
Research on Assess
Anderson, O. L., 1971. Collaborative
on the Lake Powell Region,
ment of Man's Activities
National Needs
N. S. F., Research Applied to
-
(RANN).
Water Chemistry Survey of
Bureau of Reclamation, 1965.
CHE-46, Chemical
Boulder Basin, Lake Mead,
Colorado.
Engineering Branch, Denver,
Water Quality Study of Lake
Bureau of Reclamation, 1970.
Chief Engineer
CHE-70, Office of
Mead, Report No.
Denver, Colorado.
Wash
Effect of Las Vegas
Bureau of Reclamation, 1971. The
in
Lake
Mead,
Water Quality
Effluent Upon the
Denver, Colorado.
of Zooplankton Grazing
Carlson, J. S. 1968. The EffectPhD Dissertation,
on Primary Productiviey, Davis.
University of California,
Utilities of Ecologic
Chen, C. W., 1970. Concepts and
Engineering Division.
Model, J. of Sanitary
Proposed Ecologic Model
1968. A
Chen, C. W., and G. Orlob,Environment,
Report to FWPCA by
for a Eutrophying
Water Resources Engineers, Walnut Creek,
California.
1 46
147
John Wiley and
ling Techni ues,
Sam
1963.
G.,
Cochran, W.
Sons, Inc., New York.
and N. A. Younggren, 1966.
J.
McCauley
L.,
W.
Cockrum, E.
Company.
Ei2121 /, W. B. Saunders
Study
Simulation Through a
at
A
Look
&
Cole, C. R., 1967.
Dynamics, AEC Research
Population
on Plankton
BNWL 485.
Development Report
Desirability
Clymer, 1966. The
S.,
and
A.
B.
Davidson, R.
Simulating Ecosystems, Annals
of
and Applicability of
Sciences, Vol. 128.
New York Academy
1970.
R. V. Thomann,
O'Connor,
and
Populations in
DiToro, D. M., D. J.
Phytoplankton
Bronx,
College,
A Dynamic Model of
Manhatten
Waters,
June,
Natural
New York.
of the
Selected Features
Ouri,
1958.
Techniques
Doty, M. S., and M.
Isotopic Carbon Primary Productivity
Mer., 144.
Rapp. Cons. Explor.
Report on Pollution
1967.
Environmental Protective Agency-,
Bay, Cincinnati,
and
Las
Vegas
in Las Vegas Wash
Ohio.
Biological Studies
1968a.
Environmental Protective Agency,Tributaries of the Colorado
and
of Selected Reaches
River, Cincinnati, Ohio.
1968b. Report of the
Water
Quality Criteria, D. C.
Environmental Protective Agency,
on
Committee
Office,
Washington,
National
Government
Printing
U. S.
Affect1971a. Pollution
The Lower
Environmental Protective Agency,
Lake Mead, and
Denver
Colorado, and
ing Las Vegas Wash,
Center,
Colorado River, Denver
California.
San
Francisco,
Region IX,
Mineral
1971b. The Basin,
River
Agency,
Environmental Protective in
the Colorado
and
Man-Made
Conditions
Quality Problem
A--Natural
Appendix
Affecting Mineral Quality.
Mineral
1971c. The
River
Basin,
Environmental Protective Agency,
in the Colorado
Impacts.
Quality Problem
and Economic
Appendix B--Physical
.
1 48
Environmental Protective Agency, 1971d. The Mineral
Quality Problem in the Colorado River Basin,
Appendix C--Salinity Control and Management
Aspects.
Mineral
Environmental Protective Agency, 1971e. The
Quality Problem in the Colorado River Basin,
Office
Summary--Meeting of Region IX Enforcement
of Las Vegas
Representative with Representatives
Sources.
Valley Munitipal and Industrial Waste
of Las
Everett, L. G., 1971. A Cause-Effect Analysis
Department of
Vegas Wash Effluent, Course Paper,
Hydrology, University of Arizona.
Qashu, 1971. Micronutrients and
Everett, L. G., and H. K.
to the
Biological Patterns in Lake Mead, Report
of Hydrology and
University of Arizona, Department
Water Resources Administration.
H. K.
R. D. Staker, and
Everett, L. G., J. S. Carlson,
Biological
Study of
and
Chemical
A
Qashu, 1970-71.
Canyon Section, Report 1
River--Grand
National
the Colorado
of the Interior,
Department
the
and 2, to
Canyon National Park.
Park Service, Grand
of Primary Productivity
In:
Goldman, C. R., 1963. Measurement
Freshwater with 14 C,
Factors
in
and Limiting
Primary Productivity Measurement,
Proc. Conf. on
of Hawaii.
Marine and Freshwater, University
Misuse
1966. The Use and
Shapiro,
Hahn, G. J., and S. S.
Regression, Industrial Quality
of Multiple
Control, October.
Lake
Mead
Studies,
UWater-Loss
Investigations:
298.
Paper,
Harbeck, G. E., 1958.
Geol. Survey Prof.
U. S.
Lake Mead, USA,
Jonez, 1971.
Hoffman, D. A., and A. R.
Interior, Bureau of
U. S. Department of the
Reclamation.
, Vol. I
on Limnolo
Treatise
A
Hutchinson, G. E., 1967.
and Sons, Inc.
and II, John Wiley
1965.
G. L. Oakland,
and
River
Basin,
Hembree,
Upper Colorado
Iorns, W. V., H. C.
of
the
Water Resources
S. Geological Survey,
Prof. Paper 441, U.
Washington, D. C.
149
Purdue Industrial Waste Conference, in
Kerr, P., 1970.
Technology, September.
Environmental Science and
Chemical Corp. in
Kuentzel, L. E., 1969. Wyandotte Technology, September.
Science and
Environmental
in the Symbiotic
Lange, W., 1967. Effect of Carbohydrates
Green Algae with
Growth of Planktonic Blue
Science and
Bacteria, in Environmental
Technology, September.
Eutrophication Information Program,
Lee, G. F., 1970.
Water
Paper No. 2, The University of Wisconsin
Resources Center.
197 1 .
Region
Comprehensive
Framework
Study,
Lower Colorado
Pollution Control and
Water
Quality,
XV,
Appendix
Colorado Region
Health Factors, Prepared by Lowerfor the Pacific
State-Federal Interagency Group
Southwest Interagency Committee.
of the Freshwater
Lund, J. W. G., 1965. The Ecology 40.
Phytoplankton, Biol. Rev.
Report on the Fishery
Service,
Moffet, J. W., 1943. A Preliminary
and Wildlife
of Lake Mead, U. S. FishStanford University,
Natural History Museum,
California.
Data Programs for
of
Moore, S. F., 1972. The Design
ASCE #1574, January.
Aquatic Environments,
1949. Interdivisional
on
National Research Council,
Density Currents, Subcommittee
on
Committee Lake Mead Density Currents Investiga3,
Lake Mead,
1, 2; 1941-46, Vol.
Vol.
1937-40,
tion,
Reclamation
S. Bureau of
Washington, U.
Winston,
Holt, Rinehart and
Odum, E. P., 1963. EsIllasy,
San Francisco.
Light and
of Increasing
Effects
Diatom Communities
Patrick, R., 1971. The
on the Structure of
Temperature
16 (2).
in L & 0, Vol.
and
Ecological Systems Analysis
Transactions of the
Patten, B. C., 1969.
From
Science,
98, No. 3.
Fisheries
Society, Vol.
American Fisheries
.
15 0
Fresh-Water Invertebrates of the
Pennak, R. W., 1946.
Company.
United States, The Ronald Press
Nutrition and Eutrophication,
Provasoli, L., 1969. Algal
Academy of Science.
In: Eutrophication, National
Photos nthesis,
Rabinowitch, E., and Govindjee, 1969.
John Wiley & Sons, Inc.
R. F. Kilmartin, 1971.
Riesbol, H. S., W. H. Minkley, and for the Navajo
Lake Powell Quality Studies
Water Resources Engineering
Plant, ASCE National
Meeting, Phoenix, Arizona.
Model of Regional
Riley, G. A., 1965. Mathematical
Limnol.
& Oceanog. 10.
Variations in Plankton,
Ecology
1946. Quantitative
Riley, G. A., and D. F. Bumpus,
of the Western North Atlantic,
of the Plankton
12 (3).
Bull. Bingham Oceanog. Coll.
of Eutrophication
Rodhe, W., 1969. Crystallization
Europe, In: Eutro hication,
Concepts in Northern
National Academy of Sciences.
uatic Productivit
Russel-Hunter, W. D., 1970. A
Macmillan Company, New York.
of Lakes by Agricul
Fertilization
Wks.
1947.
Wat.
Sawyer, C. N.,
J. New Engl.
tural and Carbon Drainage,
Assoc., 61.
of the Lower
Reservoirs, M.S.
Slawson, G. C., 1972. Water Quality
Effect of
Colorado River and the Department of Hydrology,
Thesis Submitted to the
University of Arizona.
1948.
Cummings, G. B.,
C.
P.,
and
in
Lake
Smith, W. O., Vetter,
Sedimentation
Comprehensive Survey of of the Interior, Geol.
Mead, U. S. Department
Reclamatio n.
Survey, Bureau of
in
Some Theoretical Problems
Notes on
R. Goldman (ed.), Primary
Steele, J. H., 1965.
in C.
Ecology
Production
Aquatic Environments, Mean. Inst.
Press, Berkeley.
Production in
California
of
Univ.
Idiobiol.,
Over Lake Mead,
Wind Patterns
1953.
22.
U. S. Weather Bureau,
Weather Bureau Tech. Paper
U. S.
151
Aquatic
Vollenweider, R. A., 1971. Primary Production in
Environments, IBP Handbook No. 12, Blackwell
Scientific Publications, Oxford and Edinburgh.
In: R. A.
Wiessner, W., 1962. Inorganic Micronutrients,
of Algae,
Biochemistry
Lewin (ed.), Physiology and
Academic Press, New York.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement