W

 

ESTIMATION OF BOTTOM AREA
IN AN. ARTIFICIAL STREAM.
USING PHOSPHORUS-32

Thesis for the Degree of M. S.
MiCHlGAN STATE UNIVERSITY
IOHN PAUL GIESY JR.
197 1

“J LEBRARY

Michigan Stab:
University

 
  
    

   
     

 
  

a; alumina BY» ‘5‘ ‘
HERE & SflNS' y
808K BINDER! INC. 7

LIBRARY BINDERS

ESTIMATION OF BOTTOM AREA IN AN
ARTIFICIAL STREAM, USING
PHOSPHORUS-52

by
JOHN PAUL GIESY jr.

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Fisheries and Wildlife

1971

1'.I‘.]n _~ HI!) “fill

in epilitlic elynl comnunity, coznOos M mainly of
dietoms, was allowed to colonize a lnborfitory stream, 21.9
meters in length. & phosphorus—32 tracer pulse of 7.721107
cpm was added in an attempt to determine the colonizable sur-
face area of the stream by 52F sorption.

Glass substrate placed in the stream removed an averepe
of 1677.7 0pm or 87.58 cpm per cmg. ’1thouyh the weight of
alfiae on the czrtii‘ic iel substrate varied from one part of
the stream to another, the activity collected by the sub—
stratn did not very "ignificnntly. This indi ted t} the
streau bottom was uniformly covered end that the 32? sorption

was a surface phenomenon, not effected t3 the swtaniin cron

of algae, during a short contnct time in the lotic 3y tem.

. - , ()1
92? added, only b.09XlO cpm or

Cf the 7. 2’107 cnm of
ap roximetely 12.5% of the init in] activity was removed. This
coupled vith 2 large eaount of variation between the w ter
ples t 'iken from the stream, made it imloesible to calcu-
late an accurnte change in activity from the Leginnin; to the
end of the stream. To be able to calculate v chen;e in activ—
ity and hence the surface area, lens activity Fist he used in

any further experiments.

ACKNOWLEDGEMENTS

I wish to express gratitude to my graduate committee,
Dr. Robert Ball, Dr. Clarence McNabb, and especially the
chairman, Dr. N.R. Kevern.

Thanks are also due to Dr..Walter Conley for his help
in processing and statistical analysis of data.

I would also like to express my appreciation to David
Jude, John Janssen, Lou Helfrich, David Rosenberger, and
James Seelye for their help in data collection.

Thanks also to my wife Susan for her patience and help
in preparation of the manuscript.

Financial support for this research was supplied by
Federal Water Quality Administration grants 5Tl-WP109 and
ITfi-WP246.

Use of Michigan State University computing facilities
was made possible through support, in part, from the National

Science Foundation.

iii

TABLE OF CONTENTS

INTRODUCTION . . . .

ARTIFICIAL STREAM
Stream bed
Reservoirs
Lighting . .
Temperature control
Water and nutrient medium
Alkalinity . .
Hardness . .
pH . . .

Phgsphorus .
Stream colonization

DETERMINATION OF BOTTOM SURFACE AREA

Tracer . . . . .
Water sampling . . . .
Algal sampling . .
Detection of activity in samples .
Weighing . .
Calculation of activity in tracer pulse
Calculation of bottom surface area
RESULTS . . .
Activity sorbed by algae . .
Dry weight . . . . .

Organic matter
Relationship of activity to dry weight
organic matter .

Back calculation of total phosphorus removed.

SUMMARY . . . . . . .
LITERATURE CITED . . . . .

iv

and

.‘
-‘J
0

Artificial stream dimensions and capacities .
Nutrient medium . . . . . . .

Algal genera present in the artificial stream .

Animal {reups represented in the artificial
stream at the time of the experiment. . .

Mean activity per slide in each trough with the
estimated standard error and coefficient of
varience for oech trough. The been sctivities u
compared by LS}. . . . . . .

Penn dry weirht of alrae per seppling slide with
stundard errors end coefficient of v~rietion .

Correlation matrix of intercorrelitions between
the dry weight, organic wetter, distance from
source and activity of eech slide . . .

Teen weirht of organic vet'er on ereh samplinp
slide with standard error and coefficient of
variation . . . . . . . .

are
50

S4

10.

ll.

1.:
IX)
0

1'50

LIST OF FIGUHgS

Page
Diagram of artificial stream. . . .
Photographs of artificial streem: (a) complete
artificial stream; (b) insive of stream trough,v ‘
showing plastic lining and vertical cylinders 8
Photographs of the apparatus for crtchin. jLP
after it passed through Jhe stream: (as auxiliary

reservoir-and pump; (b) polyethylene holdinh
tanks 0 o O O O O 0 ° 11

Diurnal temper ture fluctuation of stream
water . . . . . . . . 14

Total slkalinitv of the stream meter for the two
months preceding exoewimentation. . . 18

Free carbon dioyide concentrations for the two
months preceding caperimentation. . . 21

Diurnal Carbon dioxide cycle for water in the
reservoir just under the return pipe iron tee
stream, on 21 June 1971. . . . .

P.)
‘w

Total dissolved orthophosphete in the stream

wvter for the two months prec ding emperimen—
138131011 o o o o O o O 0 2L)

Photorrrph of the epilithic elyal community and
associated Tendipodidee l rvee . . .

In
u:

Fhotoyrvqflu or (a) {JVMNTP delivrqulxyttle; (h)

Person tnkin: a water Humyle. . . 55
Weduced verrfixurs of the lnr e {lvudns used to
calculate the activity in the irncn- pulse at the

end of ehch troueh . . . . . . 41
Activity in tracer pulse at the end of neck

trough . . . . . . . . [H5

The activity per slide as a function of

distance from the stream source with linoer
regression laust sen res line (Y=1544.t9+l4.441).

The regression line was not siynificant (C.Ol¢F<C.

2 5) 0 o o o o o o o 0 LR

vi

Firure
14. Dry weipht as a function of distance from the

source 0 O O O O O O O

15. Kean dry weinht of alrae p1. sampling slide
in each trough . . . . . .

16. Organic matter as a function of distance from
the source. . . . . . .

17. Mean .eiyht of orranic matter per sl de in each
trougtl . . . . . . . .

18. Activity as a functi n of dry weight . .

19. Activity as a function of organic matter .

INTRODUCTION

Productivity and standing crop of aquatic systems have
always been important limnological parameters, receiving
much attention. Traditionally, periphyton has been measured
in terms of weight of algae per unit area because there was
no way to determine the total colonizable bottom surface
area of lotic systems. This relative measure is useless
when calculating total production or total available biomass.

A rapid method for determining the bottom surface area
and standing crop of small rocky streams, using 32P, pro-
posed by Nelson et a1. (1969) found that the colonized bottom
area was more than four times as great as expected from calcu-
lation from surface measurements.

Studies by Reigler (1956) showed that 52F is rapidly
taken up by the algae. It has also been demonstrated that 52F
did not occur in higher trophic levels until several days
after its addition to aquatic systems (Ball and HOOper, 1961;
Davis and Foster, 1958).

It has also been found that only a small amount of 52F
is sorbed into sediments and detritus (Nelson et al., 1969).

The rapid uptake of tagged phosphorus, especially in
lotic systems where phosphorus is limiting, and the fact that
little activity is lost initially to anything other than the
epilithic algae, makes 52P sorption a very attractive method

for determining the amount of bottom area colonized by epilithic

algae.

Autoradiography of 65Zn sorption by an epilithic algal
growth has shown that nutrient uptake in a flowing system
with short contact time is a surface phenomena, involving
only the upper surface of the algal mat (Rose and Cushing,
1970).

The object of this investigation was to devise a method
to determine the colonized bottom surface area of a laboratory

stream with a known bottom surface area.

ARTIFICIAL STREAM

Artificial laboratory streams have been used for
controlled studies of various stream parameters. Artificial
streams have been used to study primary productivity (Moln-
tire et a1., 1964; NcIntire and Phinney, 1965 and Kevern and
Ball, 1965). Artificial streams have also been used to study
metabolism of a microcosm (Odum and Hoskins, 1957) and inter-
actions among stream organisms (Whitford et al., 1964). Model-
ing of aquatic pOpulations has been done using artificial
streams (Lauff and Cummins, 1964). Artificial streams have ev-
en been used as a hatchery for insect specimens (Feldmeth, 1970).

Artificial streams have been built in a variety of forms
to meet the requirements of many types of experiments. because
of the need for a relatively long stream to allow for phos—
phorus sorption, the stream used in this study was much longer
than most previous artificial streams.

Stream bed. The stream was constructed as six, galva-

 

nized steel troughs, 3.657 m long, 20.52 cm wide and 12.70 cm
deep (Table 1). These troughs were supported on an angle iron
frame such that water from one trough flowed over a 1.91 cm
baffle and drOpped 17.78 cm into the next trough (Figures 1
and 2). To conserve space the water flow was alternated from
right to left from one trough to the next, forming a total
stream length of 21.94 m.

The inside of each trough was lined with heavy, black

polyethylene with 1218 plexiglas cylinders each 1.9 cm high

5

4

Table 1. Artificial stream dimensions and capacities.

 

 

STREAM SURFACE AVAILABLE FOR COLONIZATION

 

Bottom area 449600 cm2
End area 155 "
Side area 15.600 "
Inside of baffles 387 "
Outside of baffles 619 "2
59,400 cm
Area of Plexiglas cylinders 10.200 0m2

 

TOTAL surface area available for colonization 69,600 cm2

 

Trough length 5-65 m
Trough width 12.70 cm
Total length . 21.94 m
Water depth) 5.18 cm
Cross sectional area of water in stream 40.59 cm2
Water velocity 2.69 m/min
Rate of discharge 20.00 liter/min
Reservoir volume 155.8 liters
Stream volume 458.6 liters

Total volume 594.6 liters

- .— -—-—-_.——_—._——. ...—.-_ ——-

Figure 1. Diagram of artificial stream.

.6233.
.203 .2 a...
.85.... J 0.300

50.000 .2 2.5 $008.32";
2.5 .25 .5 £30 23:932.:

 

 

02¢ 52.x ._ 3:9.»

2:3. .1 2:2“.
«553.260

F H _ h L

H 4 g _ _

co. 2. 3 on o

ddo'o'ujufio'

I'll Sunfll Tl gall!

3“; ozu

 

«>420 138% to...» 3m; we;

 

o

 

 

o

 

 

 

 

 

 

{H

 

I...
II

 

 

 

 

 

 

 

 

 

 

 

 

 

IOOCCOO- 000......

 

I"

Figure 23 livftographs <d?:artjiicied. stream:
(a) complete artificial streau; (b; inside of stream

trouqh, showing plastic lining and vertical cylinders.

ska .-
- .. 0 . 5n"

.r I... '- I" I". ‘~_
1." L559)“; . 1,}

N

'J‘_' u

' .‘0 W

n" 'f -_ '- r:

.. 3 ‘a, ._

l0. ' C ».
o

.,I
'

 

9
and 0.655 cm in daimeter, placed semirandomly on the stream
bottom to increase the surface area and provide some vertical
surface for algal colonization (Figure 2b).

Reservoirs. The stream was a recirculating system, with
water drawn from and returning to a series of three inter-
connected, wooden reservoirs with a total capacity of 141.2
liters. All three of the reservoir cells were sealed with
epoxy paint and allowed to dry for one week. After drying,
the reservoirs were flushed with running water for a day be-
fore being placed in the stream system. The reservoir cells
were placed under the stream bed SUpports, connected with PCV
pipe and covered with black polyethylene to reduce evaporational
loss and keep the water from being contaminated with zinc from
the water that condensed on the underside of the troughs.

Using submersible pumps, water was pumped from one
reservoir through tygon tubing to the stream surface (Fig-
ure 1). After circulating through the stream, water returned
to the reservoir at the other end of the series via a return
pipe from the end of the last trough (Figure 1). This kept
the water in the reservoirs from becoming stagnant while
allowing the longest time for the water to cool, after being
warmed by the lights above the stream.

The water return was built so that when 52P was run
through the stream a bypass could be connected to the exit
pipe and water containing 32F shunted into an auxiliary
reservoir (Figure 5). From the auxiliary reservoir, the water

and 52P were pumped into polyethylene holding tanks (Figure 5b)-

10

Figure 5. Photographs of the apparatus for catching

52

P after it passed through the stream: (a) auxiliary

reservoir and pump; (b) polyethylene holding tanks.

ll

 

 

 

12

Lighting. Six, 60 watt incandescent light bulbs with
reflectors and five fluorescent lamps were placed alternately
21 cm above the water in each trough, providing approximately
550 foot candles of light at the water surface. Light in-
tensity was slightly higher under the incandescent lamps than
under the fluorescents. Incandescent bulbs were used in
conjunction with grow-lux fluorescent bulbs to provide a
“balanced spectrum of light.

The 12—hour photOperiod, controlled by timers, was
from 6:00 A.M. to 6:00 P.M. The fluorescent room lights were
on constantly, resulting in a constant light intensity of
50 foot candles at the water surface during the time the
stream lights were off.

Temperature control. Water temperature was reduced to
between 12.5 C and 18.0 C with a cooling coil submerged in
the reservoir. There was a diurnal temperature fluction of
about eight degrees, due to heat input from the lights during
the twelve hour photoperiod (Figure 4). The temperature
cycle was very constant over the two months preceding experi-
mentation. The temperature, measured at 2:00 P.M. each day,
was constant at 15.5 C.

Water and nutrient medium. The stream was initially
filled with distilled water and distilled water was used to
replace water lost by evaporation.

Nutrients were added according to the Chu #10 medium,
modified by Roche as described by Prescott (1951). Micro-

nutrients not contained in the modified Chu #10 medium were

1}

Eiaure 4. Diurnal t*mperature fluctuation of stream

V'u'u’ a t er 0

1M-

u,

154

[q I d u ‘

J
5

3 .33325223

IIIII PEIIII

1'2

r2

fz

V

 

PI

TIME

15
added after Bonner (1952). EDTA (Ethylenidiaminetetraacetic
acid) was added to keep the cationic species in solution by

12 and biotin were added be-

forming complex ions. Vitamin B
cause these are the most frequently needed organic compounds
for growth of alloautotrophic algae.

Appropriate amounts of each nutrient were added initially
to make up the entire system to the concentrations required
by the combined media (Table 2). During the two months prior

12 were added to the

to experimentation, biotin and vitamin B
stream. An additional 100 g of Na(H005)2 was also added.

It was not necessary to add nutrients during the two
months prior to the tracer studies because algae was con-
tinually being sloughed off and decomposing at the bottom of
the reservoir so that nutrients were recycled (Kevern and
Ball, 1965).

Alkalinity. Total, phenolphthalein and methyl orange
alkalinity were determined titrimetrically following standard
methods (APHA, 1971). The alkalinity decreased slowly from
the time the initial nutrients were added until the time of

the tracer eXperiments (Figure 5).

Hardness. Total hardness, measured by EDTA titration,

 

using erichrome black-T as indicator, was found to be constant
at 20 ppm.

The hardness was kept low on purpose to reduce inter-
ference in determinations of the activity of water samples.
The EDTA used to keep cationic nutrients in solution may

have also reduced the measured hardness.

Table 2'

Nutrient medium.

 

 

 

 

STOCK

ADUED it t

'5 La '3‘. ‘3..:”Ll‘u

 

mUTRIEHT smear soiumipm
kgSO4'7H20* 25mg/l
Na8105* 20mg/l
fia(HCO5)2‘ lbOma/l
K2H804 5mg/1
Ferric citrate* imp/l
4.
Citric acid lmp/l
rnsou** 0.05mc/1
Knit)“. SON-f2 /1
iv- *
CuSO4'5H30 5mg/l
H . .3 g
5E03 O EmL/l
hn012'4H20** 0.4mg/l
(NH4)6M07024'4H20** ioma/l
mnmn 1.59/1
Vitamin wl’ loomo/l
liotin lm;/l

*Erescott(l9b8)

H.téonner(l952)

__ ....‘.- ._.—.- _.-.....

25m1/l
20ml/l
lOml/l

Sml/l

lOOml/l

0.5ml/l
lml/l

lml/l

lml/l

lml/l

lml/l

2ml/l
(>.55ml/1
_.HO .55ml/l

 

17

Figure 5. Total alkalinity of the str«an water

for the two months precedinv experimentation.

is

V

212:

111:

13:

3122

 

 

T ‘U I ‘ f V v '—
- -
- N

(“‘ddMUNI'IVN'IV “I101

Ila

DATE

19

pg. pH was determined, using a Leeds and Northrup pH
meter. The pH varied between 7.9 and 8.5: Host of the
variability was due to determinations being made at different
time of day. The pH varied with the amount of CO2 dissolved
in the water. I
992. The carbon dioxide concentrations in the water
were determined from the pH and total alkalinity, using
Moore's nomograph. The amount of free 002 held by the water
decreased during the two months before experimentation (Fig-
ure 6). This was probably due to the greater standing crop
of the growing community. The carbon dioxide concentration
showed a marked diurnal cycle (Figure 7). The concentration
of free CO2 decreased at the onset of the photoperiod as the
algae started photosynthesizing, until there was no free 002
in the water. After the photoperiod, the 002 concentration in-

creased to a miximum of 5.5 ppm.

Phosphorus. Total and dissolved orthophosphate was

 

determined colorimetrically by the stannous chloride method
(APHA, 1971). Total orthOphosphate was determined directly,
while dissolved orthOphosphate was determinea after filtering
water samples through a 0.45 u membrane filter. All phos-
phorus was reported as P. The phosphorus concentration
decreased from 0.20 ppm initially to 0.01 at the time of
experimentation (Figure 8). The fluctuations and large in-
crease in phosphorus between 4 June and 10 June may have

been caused by sloughing off of some algae or mixing of

the algal residue in the bottom of the reservoirs with

20

Figure 6. Free carbon dioxide concentrations for

the two months precedinr experimentation.

21'

 

112:

111:

3321

I321

 

 

I-‘II :3:

2-‘1

DATE

‘

U'

22

Figure 7. Diurnal carbon dioxide cycle for water
in the reservoir just under the return pipe from the

stream, on 21 June 1971.

24

Figure 8. Totn and dissolv d orthoehosphate in the

stream water for the two months preceding experimentation.

2 2 :1 E

.F.~

mh<o

—:~

2”—

_~-.

”*1.

-|-
'

 

5
2

...... a..._.u... .m...«._.
n.....u... .._._

 

 

._...

.-.-

(a so wdd)snuowdsowa

26
the water. The phosphorus level was constant the last three
weeks before the experiment. The phosphorus level in the
stream was about the same as the level in the Red Cedar River,
a local stream. This level was not as low as what might be
expected in a severely phosphorus limited trout stream en-
vironment, but was low enough to allow for rapid sorption but
also good colonization.

Stream colonization. The term periphyton, often used
for the algae that cover rocks and other submerged substrata,
is ambiguous, so, to avoid confusion, the encrusting algal
forms growing on stream substrata will be referred to as
epilithic algae.

Stream colonization was initiated by seeding the stream
with rocks covered with epilithic algae. Rocks taken from
the Red Cedar River; T.4N., R.2fi., NW%, SW% Section 15 Lansing
Township, Ingham County, Michigan during September l970were
placed in the stream reservoir. Within two months there was
a flocculent growth of green and bluegreen algae, where the
current did not sweep them away. Three months after seeding
the algal community had developed into a very diverse com—
munity of epilithic and non-epilithic algae (Table 5). The
diversity of the algal community decreased (Table 5) so that
by August 1971 there was a homogeneous epilithic community of
only a few species (Table 5 and Figure 9). The stream bottom
zuxi cylinders were covered with a dense film of Cocconeis
Placentula, which gave the stream bottom a distinct brown

castn. Embedded in the diatom matrix were the other two

Table 5.

27

Algal genera present in the artificial stream:

 

—————._ -_.._

Dec. 15, 1970

Apr. 21, 1971

August 5, 1971

 

Nostoc
ficenEdesmus
Ulothrix
Cbsmarium
Chlorococcum
Mugeotea
Gleocapsa
Meridion
Chlamydomonas
PEhdorina
Microcystis

Eu lena
Pinnularia
Navicula
Desmidium

lnabaena
Pemium

HO a
Chlorella

Cladophora
Rhizoclofiium

ST ro ra
Ceratium
Tribonema
Netrium

Gonatozygon
Nitzschia

Gflrosi a
FragiIaria

StephanodiScus

 

 

 

 

 

 

 

 

 

 

 

 

“.4

Chlorella

 

Gleotrichia
lhabaena

 

Nostoc
KfiEistrocesmus

 

Tibelaria

 

Mugeotea

 

Meridion

 

Nitzscia
Distoma
Scenedesmus

 

CladOphora

 

Chlorella‘
Scenedesmus*
Anabaena
Chlorococcum
Cladophora
Nostoc

zoclonium +
Choconeis placentu1a*

 

 

 

 

 

 

*Dominant species
+Patrick (1966)

 

28

Figure 9. A photograph of the epilithic algal

community and associated Tendipedidae larvae.

 

 

 

" .iiflfi .T-
'i 1'

 

 

 

 

 

 

 

 

 

 

 

30

dominant genera, Chlorella and Scendedesmus. There were

 

a few long strands of Cladophora but most of these were re-

 

moved so that they would not interfere with the determina-
tion of bottom surface area.

Besides the algal population, the stream supported a
population of protozoans and other small invertebrates (Table
4). The most numerous animals were the midges of the family

Tendipedidae, which formed tubes out of the Chlorella and

 

Scenedesmus and grazed on the green algae (Figure 9). There
were clear areas around the midge tubes where the algae had

been grazed off, down to the diatom base.

51

Table 4. Animal groups represented in the artificial stream
at the time of the experiment.

Tardigrada
Cladocera
Hotetoria
Nematoda
Protozoa
Rhnbdocoola
Flam ria
Cstracodn
Gastrotrichia
Chironominae

 

32

DETERMINATION OF BOTTOM SURFACE AREA

Tracer. Phosphorus is generally found in small quan-
tities as soluble orthophosphate in most natural, unpolluted
aquatic systems (Round, 1970). A low concentration of phos—
phorus can limit growth in most species of algae (Round, 1970).
Because phosphorus is so often limiting, many species of algae
have evolved ways of absorbing excess phosphorus when it is
available and storing it within the cell until it is needed
(Goldberg et al., 1951). It has also been found that phos-
phorus limited algae can absorb as much as 25 times as much
phosphorus as algae grown in an environment with sufficient
phosphorus (Fitzgerald, 1966).

It is this ability of algae to remove phosphorus from
water that makes phosphorus an ideal tracer to measure
water-algae interactions in a lotic system, where there is
a short contact time between algae and water.

It was thought that the colonized surface area of a small
rocky stream could be calculated by knowing the amount of
phosphorus sorbed per unit area and the total quantity of
phosphorus sorbed.

Phosphorus has been used as a tracer in phosphorus
limited streams before, but it was found that it was difficult
to detect the stable phosphorus, even when large amounts were
used (Ball and H00per, 1961).

Stable phosphorus in algae can be detected by hot water
extraction in conjunction with the stannous chloride colori—

metric method of detection. But, because of the large amounts

35

of algae required for analysis, it is difficult to use ex-
traction methods for epilithic algae (Fitzgerald, 1966).

Phosphorus has a radioactive isotope that has ex-
cellent tracer qualities. Phosphorus-32 has a single, high-
energy B- particle (1.7 Nev) and a half-life of 14.3 days,
which is long enough to work with but not long enough to make
disposal a problem (Comar, 1956). By using 52P one can obtain
a measurement of the amount of phosphorus taken up by algae
and the amount remaining in the water.

Phosphorus-52 was added from a liter, polyethylene bottle
equipped with a bottom mounted disseminator and shut-off
clamp (Figure 10a). The dissemenator was a 17 cm piece of %
inch PCV pipe that had been perforated with small holes to
allow an even addition of phOSphorus tracer across the entire
stream width, in one minute.

Three and one half uCi (7.72X107 cpm) were mixed in 1
liter of 0.001 N H01 to avoid adsorption to the sides of the
delivery bottle. One half gram of fluorescein dye was added
so that the radioactive phosphorus could be followed visually.

Water sampling. Stream water was sampled just before
the baffle in each trough. Using disposable plastic, 5 ml-
syringes several personnel removed 4 ml water samples accord-
ing to a predetermined time schedule (Figure 10b). Each
syringe contained one drop of concentrated HCl to prevent ad-
sorption of 52F to the sides of the syringe.

The initial water sample was taken as the dye front

passed each sampling station and the time recorded. The

Uipure 10. (a) Photo raph of tracer delivery bottle.

(b) jhotopraph of a person taking a water sample.

55

 

56

time was used for the samplers to keep tracx of time between
samples. Samples were taken according to the following
schedule: Samples were taken at 50 second intervals for
eight minutes or a total of 17 samples; Then three samples
were taken at one minute intervals, three at five-minute
intervals and a final sample was taken 56 minutes after the
dye front passed each station, making a total of 24 samples.

Water samples were mixed in the syringes and a sub-
sample of three ml placed in planchets with retaining circles
and dried at 45.5 C. After drying, ten drops of water and
five drops of normal hexane were added to spread the sample
evenly across the bottom of the planchet. After drying
for two hours at 45.5 C, the samples were ready for counting.

Algal sampling. Algal samp1,s were taken by colonized
three by one inch glass slides that had been placed in the
stream six months before the experiment was run (Figure 2b).
Nine slides were arranged evenly in each trough giving a total
of 54 slides in the stream. The slides were arranged so that
activity, dry weight, and organic matter could be analyzed
with a randomized block analysis of variance.

As soon as the water sampling was completed and all of
the phosphorus-52 tracer had been collected, the glass slides
were removed from the stream. The algae was scraped and
washed from each slide into a tared, 2-inch aluminum plan-
chet with a 5 cm diameter wax retaining circle in the cen-
ter to insure constant counting geometry.

It was found in preliminary experiments, that if a three

57
ml sample was added to the planchet, without a retaining
circle, the sample would dry in concentric rings very near
the lip of the planchet, causing a loss in activity and a
large amount of variability from sample to sample. Algae
was placed in the planchet with enough water to spread it
evenly in the retaining circle. After drying at 45.5 C for
one hour, ten drops of water and five drOps of normal hexane
were added to spread the algae evenly on the bottom of the
planchet for counting (Friedlander ct a1, 1956). Samples
were dried in an oven at 45.5 C for two hours before counting.

Detection of activity in samples. The activity of water
and algal samples was detected, using an end-window, gas
flow Geiger-counter, equipped aith an anticoincidence cir-
cuit to reduce background. The background, determined by
counting a series of stream water samples before the tracer
was added, was found to be 1.5 Cpm. This level was not sig-
nificant and caused no problems in counting.

Samples were counted for a total of 10,000 counts or
ten minutes. Most alyal samples were active enough to give
10,000 counts in ten minutes. Some water samples were not
this active, but most of these samples were near background
levels and did not enter into any calculations. Counting
10,000 total counts gives 2% error with.£=0.05 (Chase, 1970)

Throughout the experiment relative activities were used
instead of absolute activities because error is added with
each correction to obtain absolute activity. Background,

counter efficiency, sample geometry and backscettering were

58
all held constant.

Counter dead time caused loss of ionizing events in
proportion to the activity of each sample. The higher the
activity of the sample, the greater the coincidence loss.

Using a split standard source, the counter dead time was calcu-
lated to be:
T=1.5928x1o’6 cpm"l

The true count rate was calculated, using the following

equation.

where thrue count rate
r=observed count rate
T=counter dead time
Self absorption varied from sample to sample, depend-
ing on the sample thickness as mg/cm2. There was no correction
needed for the water samples, but the algal samples varied
enough that a selfabsorption correction factor was necessary.
Construction of a selfabsorption curve for algae was almost
impossible so the selfabsorption correction factor was cal—
culated from a selfabsorption curve for NaCl (Friedlander,
1956). This is not a perfect correction for the density of
algal cells, but is a good approximation. The correction
factor was only two or three percent of the total activity.
since 52F has a half-life of 14.5 days, the greatest
correction factor was for decay. Time of counting was re-
corded for each sample and all samples were corrected to the
time the tracer was added to the stream, using the following

equation:

59
-aT
Aone

where Aecorrected activity
A =observed activity 52
a=decay constant_£or P
(0.0000556 min )
tetime of decay

weighing. After counting, the algal samples were pre-

 

pared for weighing by drying in an oven at 45 C and a dessica-
tor alternately until there was no weight change used to
calculate the weight of the organic matter.

Calculation of activity in tracer pulse. Calculation of

 

the amount of phosphorus tracer absorbed by the algae was
calculated by subtracting the total activity leaving the
stream from the total activity added.

Calculation of the total activity in the water leaving
the stream was complicated by the fact that while passing
through the stream, the activity pulse became elongated and
diluted so that the remaining activity was contained in a
larger volume of water.

The total activity of the tracer pulse was calculated by
plotting the specific activity (chm/ml) as a function of time
(min) on large pieces of graph paper. The area under the
curve, measured with a polar planimeter, was eXpressed as
cpm/ml-min (Figure 12). By knowing the stream rate of flow
(20.0 liter/min) the total activity in the tracer pulse was
calculated. The activity of the tracer pulse was also calcu-
lated at the end of each trough.

Calculation of bottom surface area. with a few as-

 

40

Figure 11. Reduced versions of the large graphs used
to calculate the activity in the tracer pulse at the end of

each trough.

 

 

 

 

_ =3:

 

 

 

 

(lw/wdDMLIAIlDV

42

sumptions the bottom surface area can be calculated from

the amount of activity removed per unit area of colonized
surface area and the total amount of activity removed from the
water by the algae. The total activity removed divided by

the activity removed per unit of area will give the bottom
surface area.

It must be assumed that the bottom area to be determined
is completely covered with algae and there is little super-
flous filamentous algae, which will bias the estimate of
bottom area. It must also be assumed that the algae in the
study area all remove phosphorus at the same rate. It is
also assumed that the tracer is instantly and completely
mixed with the stream water, assuring constant contact with
the epilithic algal mat. The change in phosphorus tracer
concentration has been shown to cause a change in Uptake by
the algae.

The time of contact between tracer and algae and con-
centration of tracer also determines how much phosphorus is
sorbed. The greater the contact time and the concentration
of phosphorus, the greater the amount the algae are able to
remove. Since the area is calculated by the amount of phos-
phorus removed by the algal community, there is an implied
decrease in water phosphorus concentration.

Since this cannot be assumed to remain constant it must
be controlled out of the experiment. To avoid inconsis-
tancies from changing tracer concentrations the stream study

area was divided into subsections, much like is done in the

45
calculus, where smaller and smaller sections of the whole are
studied and assumed to form a continuum. Because of sampling
limitations, the smallest subsection possible was one trough.

The rate of uptake by algae is influenced by many factors.
In a natural system these factors could not be controlled, but
in an artificial stream they can be controlled out as variable
parameters. Different types of algae take up phOSphorus at
different rates (Goldberg et al., 1951). If the algal com—
munity is homogeneous, the sorption will be quite similar
throughout the system.

Nutrient uptake by algae is also influenced by water
temperature and current (Davis and Foster, 1958). Light also
affects the ability of algae to absorb phosphorus. Some
species of Chlorella have been found to absorb phosphorus
five times as fast in the dark as in the light (Fitzgerald,
1966). Light, temperature, current, and community homogeneity
were held fairly constant throughout the entire stream for

any point in time.

RESULTS

So much phosphorus was added in the present study, that
there was no measurable change in the tracer concentration
from the initial spike to the pulse leaving the stream (Fig-
ure 15). The apparent decrease in the pulse activity during
the first trough was due to a sampling error.

The variability in sampling made it appear that there
was more activity at the end of all the troughs, but the first.
The phenomena of over-calculating the activity of the pulse is
unexplained, but may be due to an uneven flow of tracer. 1f
the 52P was not evenly distributed in the water and samples
were taken from the more concentrated area an over estimation
of the total activity would occur. This is possible because
water samples were taken just under the surface to avoid tak-
ing algae in the samples. The tracer-dye solution had a
tendency to float along the surface in the first trough, but
seemed to be mixed after falling into the second trough. If
the tracer had not been thoroughly mixed the above mentioned
sampling bias could have occurred.

Activity sorbed by algae. The activity removed by the
algae on each sampling slide was quite variable (Figure 14).
There was a slight increase in the activity removed per
slide the farther the slide was from the source of the
stream (Figure 14). The least squares regression was not
very significant (0.10<F40.25).

The average amount of activity removed by the algae on

the slides in each trough seemed to increase from one
44

45

Figure 12. Activity in tracer puls at the end of

each trough.

46

1t.-

3 T
TRO UGH

i

I

 

4 41 1 i 1 1 1 < 1 1
' 5 '

mo: E3523 2. >:>:u<

47

Figure 15. The activity per slide as a function of
distance from the stream source with a linear repression
lsast souares line (Y=l544.€9+14-44X). The regression line was

not considered to be signific nt (C.lO<FCO.£5).

1+83

Emugom EOE muzfma

 

* d d‘ [‘1 4 I q 1 d‘ ‘ [1 q I

1“ —‘11 i 1! [1| 4 ‘ 4— 1 J‘ 4 q i ‘ dl—I ‘ d 41? d. ‘11—

1 11111

—.
__—.._

1' V

g a
0119 Had All/\llDV

a,
(ni‘da) a

99mm"

 

49

trough to the next (Table 5) but the variance within troughs
was so great that no difference could be demonstrated. The
variance showed a decreasing trend from one trough to the
next (Table 5). The increasing removal of activity and the
decrease in variance may be due to the fact that the activity
was better mixed in the water or there was an increased con-
tact time so the tracer pulse became elongated. The cal-
culated coefficient of varience for each trough (Table 5)
seems to indicate that there was a true decrease in the var-
iance and not just a change due to the difference in activity
sorbed.

A LSR test at the 0.05 level failed to show any differ-
ence in the mean activity removed in each trough (Table 5).
It can be concluded that there was no significant difference
in the amount of activity absorbed over the distance of the
stream. This may have been due, in part, to the fact that
there was a small decrease in activity in the tracer
pulse.

Dry weight. The dry weight of algae on the sampling

 

slides varied considerably from slide to slide (Figure 15).
Attempts to fit linear polynomial regression curves to the
data failed. There were no curves with a significant re-
gression.

Dry weight was analyzed with a randomized block analysis
of variance, blocked over troughs. No significant differ-
ence could be shown between the mean dry weights of differ-

ent troughs. There was a trend for the variance to decrease

‘_':_31(..l ; 11 .

ll. 851 (I

estimated

for

TUUUC?

SII’KIII 1

mac}:

actj.vity
tanuurd

IT I; v: m

50

per slide in
error
It? ()Uf h . The

and
ri‘ (1' ri

each

COO [ii 3
acLivitics

tiuurh

'HJitt

the

ient of V rience

 

i J -. O l

f.

(7'

1538.120pm
H

1’05.fi<

ff.)

l-3)£.70 ”
1943.?k “
lé§74.‘)l

1779.51 ”

 

r.
C.

lb();.fx’

LUH

.05

..__ - .4 -... .....—.»—-—.__—.__- —.. _ -..—-—.— -r‘ ..

\34

155.02

1:'. 9‘ o {:8

159.52

a

H))CO7O 177/0

3!

-- __ -....._.

\j".

are CCHQLW'CU 1).)"

 

b

1%:‘117'- 0.20 1C; Agoxéj

 

 

Figure 14. Dry weight as a function of dist nce from

source 0

Aevmompom EOE muzima.

P ? r b D

I bi - p 5 hr

P b P

r

F

b h b! b

h P h b h, L by I [P r

 

‘ p P F

l

. _

a

—.
:3:

0
fl

.2

T

f

(whens 83d

 

HQIHM A80

55
in the troughs farthest from the stream source, but calcu-
lation of the coefficient of variation showed that the de—
crease was mainly due to the decrease in average dry weight
per trough (Table 6). The average dry weight per trough
seemed to increase with distance from the source (Figure 15)
but this could not be shown statistically because of the
variance within troughs.

The mean dry weight per slide increased from trough two
until trough six where the mean dry weight decreased (Table
6, Figure 16). No significant difference could be shown
between troughs using a 13R test.

Organic matter. The organic matter was highly correlated

 

with dry weight (Table 7) as would be expected. The organic
matter was studied in the experiment because it was thought
that an increase in organic matter not in preportion to dry
weight would indicate an algal colony of green and blue-
green algae instead of diatoms. It was thought that this
may have an influence on 32F uptake.

The organic matter was distributed almost directly
proportional to the dry weight, which indicated that the
algal pOpulation was homogeneous throughout the entire stream.
The average organic matter per trough followed the same pat-
tern as that of dry weight with a decrease in the second
trough and a great increase in trough 5 (Figures 17 and 18).
The variance did not show any trends and there was consider-

able variation within the troughs (Table 8).

Table b.

with standard error and

Lean dry Jeight of algae per sampling
coefficient of variation.

slide

 

wafiizr‘u saw" wranglmarr

 

 

 

4?me

U1

lb.80mg
15.6%) ”
16.50 "
17.14 "
25.04 "

n

18.513
c.7b"
15.05"

 

55

Figure 15. lean dry weirht of alaae per scmpling

(

slide in e~ch trough.

 

 

2h

f

V v V V j v
_- u-
N —

(5w)lHOI 3M Auo

ll

TROUGH

Table 7.

57

Correlation matrix of intercorrelations between

the dry weight, organic matter, distance from source and

activity of each slide.

 

l 2
1.000 0.905*
0.905* 1.000
0.130 0.087

0.513 0.571

\N

0.190
O .C)(",7
1.000

O .281)

 

l=0ry weight
2=0rganic matter
5=Distance
4=flctivity

*flirnificant correlation

58

Figure 16. Organic matter as a function of dist nce

from the source of the stream.

59

"III!

*1 Y I

(Stu) 30:19 83d HallVW SINvoao

f

I

I“

f

f

DISTANCE FROM SOURCE (m)

 

60

1 ~ - .rn ° rm 9 e c‘lide
P'Wule 17. i"arm weight of organic Hatt.r p r o
l C I

in 6x301: trrngfih.

61

 

 

17+

a —
' v t a v .3
#

(5w)8311VW oleoao

62

Table ’. Venn weinht of orranic mat’er on each sampling
slide with standard error and coefficient of variation.

 

'PHCWVUi (NK15N1C115;TTWJ. S .5(lcxx‘)

l 15.9?mg 1.20my “.691

2 10.25 " 4.?4 " 41.45"
22’ 11.12 H 1.00 " 9.01”

11.92 “ 0.58 " 4.56"

U1 4:

15.64 " 2.2e." 14.+5"

6 16.66 " 2.65 ” 82-75"

 

65

Relationship of activity to dry weight and organic
matter. studies by Rose and Cushing (1970) showed that
nutrient sorption by epilithic algae was a surface phenom-
enon. Rose and Cushing were using relatively thick algal
mats of about 500-600 u. Even though the algal mats used in
the present study were much thinner, it was assumed that
the sorption of phosphorus would be a surface phenomenon.
Activity seemed to be correlated with dry weight and organic
matter (Figures 19 and 20). However, a multiple regression
correlation showed that activity was not significantly
correlated with dry weight or organic matter (Table 7). The
inability to show sianificant correlation was probably due
to the great variability in the samples. It is probably
true that the thickness of the algal mat would not affect the
nutrient uptake if the layer of epilithic algae is thicker
than some critical value. This is because if the algae on
the substratum were very thin, giving rise to uncovered places,

sorption would be decreased.

64

Figure 18. Activity as a function of dry neicht.

65

1 V 1 I

' 1'0

DRY WEIGHT (mg)

 

 

' f v

1500'

V I V

(qu 3)

vi i

. €-
AllAllDV

1500‘

Activity as

66

8

function of organic matter.

67

if h

as ”.322 22.38

h P h P u I b

h L ? I b D b D I D ’ I h D

(wda) AlIAIlDV

a...
V

.32

V V’

v
v

 

68

Back calculation of total phosphorus-52 tracer removed.
By knowing the mean activity removed per unit of area and
the total stream area, the activity of phosphorus-32 tracer
removed in the stream system was calculated.

The glass sampling substrata and attached algae removed
an average of 87.38 cmecm-2. Using this mean and assuming
the sorption onto the rest of the bottom was the same, the
calculated activity of tracer removed was 6.09X106 cpm. This
was only 12.5% of the activity added in tracer pulse.

This low coefficient of sorption coupled with the variability

of the water samples, made it impossible to calculate the

colonized bottom surface area by tracer sorption.

SUMMARY

The artificial stream functioned well as an experimental
unit. Algae colonized well and a homogeneous eplithic
algal community went through a series of successional stages
and stabilized as a diatom dominated community that com-
pletely encrusted the entire stream bottom.

Since no experiments of this type had been attempted
before, it was difficult to determine the amount of P-52
tracer to be added. The estimate of the needed amount, from
pilot studies, was high. This made it impossible to determine
the bottom surface area by phosphorus sorption because only a
small amount of the tracer was removed by the stream system.

Phosphorus sorption was found to be strictly a surface
phenomenon, not influenced by the thickness of the algal mat.

The use of phosphorus-52 as a tracer in artificial
streams is promising, not only for determining bottom area
but for determining the effects of different parameters on
phosphorus uptake by algae.

finy future experiments with 52F as a tracer must find a

way to reduce the variance.

69

LITERATURE CITED

American Public Health Association. 1971. Standard
Methods for the Examination of Water and Waste Water.
15th ed., New York. 874p.

Ball, R.C. and F.F. Hooper. 1961. Translocation of
Phosphorus in a Trout Stream Ecosystem. Radio-
ecology (Edited by Schultz, U. and A.W. Klement jr.).
Reinhold, New York. 217-228.

Bonner, James A. 1952. Principles of Plant Physiology.
W.H. Freeman 00., New York. 481p.

Chase, G.D. and J.L. Rabinowitz. 1970. Principles of
Radioisotope Methodalogy. Burgess Publishing 00.,
Minneapolis, Minn. 655p.

Comar, C.L. 1956. RadioisotOpes in Biology and Ag-
riculture-Principles and Practice. McGraw Hill
Book Co., New York. 481p.

Davis, J.J. and F.R. Foster. 1958. Bioaccumulation of
Radioisotopes Through Aquatic Food Chains. Ecology.

59: 550-555.

Feldmeth, G.P. 1970. A Lar,e Volume Laboratory Stream.
Hydrobiologica. 35(314g: 597-400.

Fitzgerald, G.P. 1966. Extractive and Enzymatic Analysis
for Limiting of Surplus Phosphorus in Algae.
Journal of Physiology. 2: 52-57.

Friedlander, G. and J.W. Kennedy. 1956. Nuclear and
Radiochemistry. John Wiley and Sons, Inc., New York,
N.Y. 468p.

Goldberg, E.D., T.J. Walker and A. Whisenard. 1951.
Phosphate Utilization by Diatoms. Biol. Bull.
101: 274-284.

Kevern, N.R. and R.C. Ball. 1965. Primary Productivity
and Energy Relationshi s in Artificial Streams.
Limnol. Oceanog. 10(1 : 74-87.

Lauff, G.H. and K.W. Cummins. 1964. A Model Stream for
-Studies in Lotic Ecology. Ecology. 45: 188-190.

McIntire, G.D., R.L. Garrison, H.K. Phinney and C.E. War-
ren. 1964. Primary Production in Laboratory Streams.
Limnol. Oceanog. 9: 92-102.

71

McIntire, G.D. and H.K. Phinney. 1965. Laboratory Studies
of Periph ton Production and Community Metabolism
in Lotic nvironments. Ecol. Monographs. 55:

Nelson D.J., N.R. Kevern, J.L. Wilhm and N.A. Griffith.
1969. Estimates of Periphyton Mass and Stream Bottom
Area, Using Phosphorus-52. Water Research. 5: 567-
575.

Odum, H.T. and C.M. Hoskin. 1957. Metabolism of a Lab-
oratory Stream Microcosm. Inst. Marine Sci. Texas.

Patrick, Ruth and C. Reimer. 1966. Diatoms of the United
States Vol.1. Monographs of The Academy of Natural
Sciences of Philadelphia. 688p.

Prescott, G.W. 1951 Algae of the Western Great Lakes
Area. Cranbrook Inst. Sci. Bull. 51: 946p.

Reigler, F.H. 1956. A Tracer Study of the Phosphorus
Cycle in Lake Water. Ecology. 57: 550-562.

Rose, F.L. and G.E. Cushing. 1970. Periphyton:
Autoradiography of Zinc-65 Adsorption. Science.
May 1 (168): 576-577.

Round, F.E. 1970. The Biology of the Algae. Edward
Arnold Lt., London. 269p.

Whitford, L.A., G.E. Dillard and G.J. Schumaker. 1964.
An Artificial Stream Aparatus for the study of Lotic
Organisms. Limnol. Oceanog. 9: 600-601.

 

MICHIGAN STATE UNIVERSITY LIBRARIES

III I; I III I

306' 3651

l l

3 1293

 

l'