WWWHHWIWWWNIHWNIUI\lHlWlMlll

   

FHE\“§I‘§
2037 UBRARY

Michigan State
University

 

 

 

 

 

This is to certify that the
thesis entitled

LAKE METABOLISM AND TROPHIC STATE IN LAKE ERIE:
EVALUATION BASED ON RIP RATIOS DETERMINED BY
THE CONCENTRATION AND STABLE ISOTOPIC
COMPOSITION OF DISSOLVED 0;»

presented by

LEAH K. PIWINSKI

has been accepted towards fuiﬁllment
of the requirements for the

MS. degree in GEOLOGICAL SCIENCES

 

 

WM a CUL—

Major Professor’s Signature

AﬁrtI (L we?
r L

 

Date

MSU is an aﬂinnative-action, equal-opportunity employer

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDUE

DAIEDUE

DAIEDUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/CIRC/DateDue.indd-p.1

LAKE METABOLISM AND TROPHIC STATE IN LAKE ERIE: EVALUATION
BASED ON R:P RATIOS DETERMINED BY THE CONCENTRATION AND
STABLE ISOTOPIC COMPOSITION OF DISSOLVED 02
By

Leah K. Piwinski

A THESIS

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

MASTER OF SCIENCE
Department of Geological Sciences

2007

ABSTRACT
LAKE METABOLISM AND TROPHIC STATE IN LAKE ERIE: EVALUATION
BASED ON R:P RATIOS DETERMINED BY THE CONCENTRATION AND
STABLE ISOTOPIC COMPOSITION OF DISSOLVED 02
By
Leah K. Piwinski

To characterize lake metabolism and trophic state in Lake Erie, the relative
importance of photosynthesis and respiration (R:P) was evaluated during August
of 2003 based on the concentration and stable isotopic composition of dissolved
Oz. 02 saturation in the epilimnion ranged from 0.8 to 1.1, indicating an inﬂuence
of atmospheric gas exchange on 02. 02 saturation values consistently less than
1.0, with a minimum value of 0.04 in the hypoxic central basin, suggested
respiration was the process controlling hypolimnetic 02. 8180-02 values
lake-wide ranged from -4.0 to 12.9 %o. This wide range in isotope values
reﬂected a strong inﬂuence of biological activity on 02 cycling relative to gas
exchange. Slightly net heterotrophic conditions were evident in R:P values for
the epilimnion suggesting that R:P may not provide a true measure of trophic
state in eutrophic systems but reﬂect temporal and spatial decoupling of primary
production and respiration. Assuming a closed system, a fractionation factor (B)
of 9.0 960 was calculated from hypolimnion samples using a Rayleigh Model. A
comparison of the calculated a value to that expected for water column (23.5 %o)
and sediment (3 960) respiration revealed that 71% of hypolimnion respiration in
Lake Erie occurs in sediments. Consequently, the development of hypoxic

conditions in Lake Erie is largely driven by respiration in sediments.

ACKNOWLEDGEMENTS

Support for this work was provided by the US. EPA Great Lakes National
Program Ofﬁce. I would like to thank the members of the Lake Erie Trophic
Study team as well as the crew of the RN Lake Guardian for their dedication and
much needed assistance during sample collection. I am truly indebted to my
advisor, Dr. Nathaniel Ostrom for his guidance and support from the very
beginning and most of all for his unwavering patience. I thank my committee
members, Dr. Peggy Ostrom, Dr. Lina Patino, and Dr. Grahame Larson for their
guidance and commitment. I would like to express my gratitude to Dr. Hasand
Gandhi, Dr. Mary Russ, and Amanda Field for Showing me the ropes inside and
outside of the lab. I would especially like to thank Malee Jinuntuya for depriving
herself of sleep for a week in order to help me collect samples. I sincerely thank
the principles and staff of Fitzgerald Henne & Associates, Inc. for their patience
and understanding throughout this much longer than anticipated process.
Finally, I would like to express my deepest gratitude to my parents, Al and Cindy

Piwinski and my sister, Sarah, for their boundless support and encouragement.

iii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................................. v
LIST OF FIGURES ............................................................................................... vi
INTRODUCTION .................................................................................................. 1
METHODS ............................................................................................................ 6
RESULTS ........................................................................................................... 10

West to East Transect .............................................................................. 10

Diurnal Study ........................................................................................... 14
DISCUSSION AND CONCLUSIONS .................................................................. 18
TABLES .............................................................................................................. 25
FIGURES ............................................................................................................ 28
REFERENCES ................................................................................................... 36

LIST OF TABLES

Table 1: Data summary for samples collected in August, 2003 ......................... 25

Table 2: R:P ratios calculated for the Lake Erie epilimnion ................................ 27

LIST OF FIGURES

Figure 1: Locations of sampling stations forming a west-east transect in Lake
Erie. The coordinates and maximum depth of each station are as follows: 91M
(41 50.4°N, 82 55.0°W; 9 meters), 43 (41 47.3°N, 81 56.7°W; 21 meters), 78M
(42 07.0°N, 81 15.0°W; 22 meters), and 15M (42 31.0°N, 79 53.5°W; 62

meters) ............................................................................................................... 28

Figure 2: 02 saturation determined from CTD measurements plotted versus 02
saturation determined from Winkler titrations. The line represents a 1:1
relationship. Open diamonds are epilimnion samples for all stations, grey
squares are metalimnion samples from stations 43 and 78M, solid triangles are
hypolimnion samples from stations 43, 78M, and 15M ....................................... 29

Figure 3: Temperature (°C) and chlorophyll-a ﬂuorescence (RFU) as a function
of depth along a west-east transect (91 M - 15M). The maximum depth at each
station is represented by a hatched line ............................................................. 30

Figure 4: 02 saturation (data obtained by Winkler titration (squares) and CTD
(gray line)) and 6180-02 as a function of depth along a west-east transect

(91 M - 15M). The maximum depth at each station is represented by a hatched
line ...................................................................................................................... 31

Figure 5: 02 saturation plotted versus time for samples collected over a 24-hour
diurnal cycle at central basin stations 43 and 78M. The shaded area represents
nighttime conditions. Samples were collected at 5m(diamonds), 12m(squares),
16m(triangles), and 20m(circles) at station 43 and 5m(diamonds), 15m(squares),
18m(triangles), and 20m(circles) at station 78M ................................................. 32

Figure 6: 8180-02 plotted versus time for samples collected over a 24-hour
diurnal cycle at central basin stations 43 and 78M. The Shaded area represents
nighttime conditions. Samples were collected at 5 m(diamonds), 12m(squares),
16m(triangles), and 20m(circles) at station 43 and 5m(diamonds), 15m(squares),
18m(triangles), and 20m(circles) at station 78M ................................................. 33

Figure 7: 6180-02 plotted versus 02 saturation for stations 91M(Circles), 43(open
squares), 78M(open diamonds), and 15M(triangles) .......................................... 34

Figure 8: 8180-02 plotted versus 02 saturation reported for Lake Erie 2003

(triangles), Lake Erie 2002 (squares) (Ostrom et al. 2005), Grand Traverse Bay
(open diamonds) (Field, 2004), and Lake Superior (circles) (Russ et al. 2004).. 35

vi

INTRODUCTION

An understanding of the factors controlling lake metabolism and trophic
state in Lake Erie has been sought after for over 30 years. Excess loading of
phosphorus from anthropogenic inputs in the 1960’s caused algal blooms and
eutrophication of the lake (Rosa and Burns, 1987; Makarewicz and Bertram,
1991; Bertram, 1993; Sweeney, 1993). As a result, oxygen consumption by
bacterial respiration increased, promoting hypoxic conditions in bottom waters
and threatening ﬁsheries (Bertram, 1993; Sweeney, 1993). Hypoxia is deﬁned
as oxygen concentration less than 2 mg/L (z 62 umol/L) (www.epa.govlglnpo/
glindicatorslwaterloxygena.html). In response to the creation of legislative and
restoration programs aimed towards reducing nutrient inputs from industrial,
agricultural, and wastewater sources, conditions in the lake have signiﬁcantly
improved since the early 1970’s (Makarewicz and Bertram, 1991; Sweeney,
1993). In recent years however increased phosphorus concentrations and
increased frequency and extent of hypoxia in hypolimnetic waters have been
observed (Charlton and Milne, 2005). In this paper we evaluate the cycling of 0;
through concentration and stable isotope determinations. In aquatic ecosystems
02 is cycled via gas exchange with the atmosphere and the biological processes
of photosynthesis and respiration. The balance between primary production and
respiration is often termed lake metabolism, which has been shown to vary with
the trophic state of aquatic environments (del Giorgio and Peters, 1993; 1994).
External inputs of organic carbon tend to dominate lake metabolism in

oligotrophic systems resulting in an excess of respiration (net heterotrophy)

whereas in eutrophic systems primary production predominates over respiration
fueled by both internal and external carbon inputs (net autotrophy). Thus an
evaluation of 02 cycling by primary production and respiration is perhaps the
most fundamental measure of ecosystem function and trophic state. The primary
objective of this study was to evaluate lake metabolism and trophic state in Lake
Erie utilizing the concentration and stable isotopic composition of dissolved 02
([02] and 5180-02) and the ratio of respiration to photosynthesis (R:P).

The processes that control oxygen ﬂux in an aquatic ecosystem are
photosynthesis, respiration and atmosphere exchange (Bender and Grande,
1987; Quay et al. 1995). A standard measure of the relative magnitude of
respiration and photosynthesis is the fraction of 02 saturation; the deviation of
the measured 02 concentration from that expected when the water column is in
equilibrium with the atmosphere at a particular temperature (Weiss, 1970). 02
saturation values equal to one indicate equilibrium with the atmosphere at a
given temperature, values greater than one indicate oxygen supersaturation and
values less than one are indicative of oxygen undersaturation. Therefore,
undersaturation would be Characteristic of net 02 consumption (respiration) while
supersaturation is indicative of net 02 production (photosynthesis).

The relative importance of photosynthesis, respiration and atmospheric
exchange cannot be determined from 02 concentrations alone. Additional

information, however, can be gained from stable isotope data (Bender and

Grande, 1987). Values of 5130-021 are expressed with respect to 02 in
tropospheric air in which 02 derived from the atmosphere has a value of 0.0 960.
There is a slight isotopic fractionation effect during dissolution that favors the
heavier isotope and shifts the value for dissolved 02 to 0.7 960 (Knox et al. 1992).
During photosynthesis 02 is derived from the water molecule without fractionation
(Stevens et al. 1975; Guy et al. 1993) and therefore the 02 produced is similar in
isotopic composition to the surrounding water. In Lake Erie, the 8180 of water
and 02 produced by photosynthesis is approximately -29.7 96° (Ostrom et al.
2005a)

A substantial decline in the isotopic composition of 02 occurs during
periods with high rates of primary production (Stevens et al. 1975; Bender and
Grande, 1987). During respiration, the consumption of oxygen leaves the
residual 02 pool enriched in 18O due to preferential uptake of 160 (Lane and Dole,
1956; Bender and Grande, 1987; Kiddon et al. 1993). Therefore, 8180-02 values
greater than those reﬂecting atmospheric equilibrium, 0.7 %o, are indicative of
respiration; while values less than 0.7 960 are the result of production from
photosynthesis (Bender and Grande, 1987). If steady state is assumed, such
that exchange with the atmosphere equals net production the R:P ratio can be
determined from 6130-02 and [02] using the equations presented by Quay et al

(1995).

 

' Oxygen stable isotope ratios are expressed in per mil (360) notation: 8‘80 = [(RmdeWII] x
1000. 618002 values are expressed with respect to tropospheric Oz. 02 in air has a value of
23.5 %o with respect to VSMOW (Vienna Standard Mean Ocean Water).

Data on rates of primary production in Lake Erie are few which likely
reﬂects the analytical limitations of traditional incubation approaches
(Glooschenko, 1974; Ostrom et al. 2005a; b). The R:P ratio, however can be
readily obtained independent of rate measurements using the concentration and
isotopic composition of 02 (Ostrom et al. 2005b). Thus insight into lake
metabolism of an ecosystem can be obtained across broad spatial and temporal
scales (Quay et al. 1995; Field, 2004; Russ et al. 2004; Ostrom et al. 2005b).
Eutrophic lakes are generally considered net autotrophic whereby a
predominance of in situ primary production causes photosynthesis to exceed
respiration (R:P<1) (Jones, 1992; del Giorgio and Peters, 1993; 1994; Cole,
2000). In oligotrophic lakes the input of organic carbon from the watershed
results in excess respiration relative to photosynthesis creating a net
heterotrophic system (R:P>1) (Jones, 1992; del Giorgio and Peters, 1993; 1994;
Cole, 2000). Use of R:P to assess trophic state has been validated for a variety
of lake ecosystems where it correlates with traditional measures including
Chlorophyll-a and total phosphorus concentrations (Jones, 1992; del Giorgio and
Peters, 1994).

Recently the use of R:P as a measure of trophic state has been called into
question as periods of net autotrophy have been observed in net heterotrophic
lakes (Hanson et al. 2003; Field, 2004; Russ et al. 2004; Ostrom et al. 2005b).
The debate has been fueled by the role of DOC concentrations and sediment
resuspension events in lake metabolism as well as methodological limitations

(Carignan et al. 2000; Hanson et al. 2003). The observation that lake

ecosystems may ﬂuctuate between periods of autotrophic and heterotrophic
conditions was recently observed in Lake Superior (Russ et al. 2004) and Grand
Traverse Bay in Lake Michigan (Field, 2004). For these predominately
oligotrophic environments, periods of net autotrophy observed in the spring were
the result of temporal decoupling of primary production and respiration and that
the observation of net autotrophy within oligotrophic systems challenges earlier
views of R:P as a trophic state indicator (Field, 2004; Russ et al. 2004; Ostrom et
al. 2005b). This study was conducted in Lake Erie, a known eutrophic system, to
further evaluate the use of oxygen isotopes and R:P to Characterize trophic state

and better understand the factors contributing to lake metabolism.

METHODS

Samples were collected in Lake Erie during August of 2003 aboard the
EPA research vessel Lake Guardian. Four open water stations were selected
along a west-east transect (91 M, 43, 78M, 15M) that represent a gradient of
trophic conditions from the eutrophic western basin to mesotrophic eastern basin
(Figure 1). Water samples were collected using 8 L, lever-action Nisken bottles
attached to a rosette, triggered at depths representative of the epilimnion,
metalimnion, and hypolimnion throughout the thermally stratiﬁed water column.
Temperature, chlorophyll-a ﬂuorescence, and 02 concentration measurements
were obtained at half-meter intervals using a Seabird 25 CTD proﬁler deployed
within the rossette. The concentration of 02 was also determined from samples
collected in 300 mL BOD bottles using a modiﬁed Winkler titration method
(Carpenter, 1965; Emerson et al. 1999). To evaluate the precision, with which
titrations were performed, replicate 02 concentration samples were collected at
station 43 from depths of 5 and 16 meters. The standard deviations of replicate
02 concentration data were 3.05 and 1.10 pmol/L for the 5 and 16 m samples,
respectively.

02 saturation (expressed as a fraction) in Lake Erie was calculated using
02 concentration data obtained by Winkler titration and CTD temperature
measurements (Weiss, 1970). At stations 91M (western basin) and 15M
(eastern basin) saturation values from the CTD proﬁle corresponded well with
those determined by Wrnkler titration at the same depths (Figure 2). In the

central basin Oz saturation determined by Winkler titration was less than that

evident within the CTD proﬁle at the oxycline. This lack of agreement between
the CTD and titration proﬁles likely resulted from the tendency for the Niskin
bottle to collect water from a depth range corresponding to the height of the
bottle (z1 m) whereas electrode measurements are obtained over a much
narrower interval. As the CTD reﬂects changes occurring over narrower depth
intervals than bottle measurements, we use this data in discussion of variations
in 02 concentrations. 02 saturation determined by Winkler titration, however,
was used in the calculation of R:P given that both [02] and 6180-02 samples were
evaluated from the same water collected within a Niskin bottle.

Collection and preparation of samples for determination of the 6180-02
followed the procedure of Emerson et al. (1991, 1999). Prior to sampling, 1 mL
of saturated mercuric chloride solution (HgCl2) was placed within 200 mL glass
vessels equipped with high vacuum stopcocks and dried at 90°C to halt biological
activity upon sample introduction (Emerson et al. 1999). The glass vessels were
then evacuated and the necks ﬁlled with distilled H2O to prevent contamination
from air. Sample collection involved slowly purging the bottle neck with CO2 and
allowing a stream of water from the Niskin to overﬂow the neck while gradually
opening the stopcock. The vacuum draws lake water into the vessel where upon
02 degases into the headspace. Approximately 100 mL of lake water was
collected in each vessel. Prior to analysis, samples were stored for 4 to 7 weeks
with distilled H20 in the necks of the glass vessels. In the laboratory, samples
were eqUilibrated by submerging and rotating them in a water bath kept at a

constant temperature of approximately 24°C for at least 6 h. Prior to 6'80

determination, all but 1 mL of lake water was removed from each vessel using a
vacuum pump.

8180-02 was determined by gas chromatograph-isotope ratio mass
spectrometry (GC-IRMS) as described in Roberts et al. (2000). The sample
vessel was attached to the evacuated inlet of a gas chromatograph where
ascarite and LiOH was used to remove CO2 and water upon the release of
sample gas into a pre-evacuated 3 mL gas sampling loop. Helium was used as a
carrier to move the gas to a 60 column. A molecular sieve 5A, 8 m long column
was used to separate N2 and 02 gas in time. 02 gas was allowed to enter the
mass spectrometer where the isotope ratio was determined relative to a
previously equilibrated reference gas. GC-IRMS provides an analysis of the
isotopic composition of 02 with a precision of :03 960 or better (Roberts et al.
2000)

R:P ratios were calculated from measured [02] and 5180-02 using the

equations presented by Quay et al (1995).

(1) R'P = (18116pr _ 18/1509) / (18,160ar _18/1609)
where

(2) 18/1609 = docs/160306 " (lozlsol/ [Ozlsat)1al160}l{1 - ([02130l/[02]sm)}

The isotopic composition of H20, atmospheric O2, and dissolved 02 are
represented by the terms 18’ 16CW, 18"60,. and 18“‘50 respectively (Lane and Dole,

1956; Quay et al. 1995). 18"60 is the measured 8180-02 value determined for the

sample. 18’ 16OW, measured by Mountain Mass Spectrometry, was determined to
be -29.7 %o (with respect to air) for Lake Erie water (Ostrom et al. 2005a) the
isotopic composition of the atmosphere (‘8’ 160,.) is 0 960. 18“609 is the calculated
gas transfer ratio of net 02 ﬂux between the water column and the atmosphere,
which varies depending on the measured concentration of 02 (Quay et al. 1995).
[02]”. is the measured concentration of O2 dissolved in solution and the [02],,“ is
the expected concentration of dissolved 02 in equilibrium with the atmosphere at
the observed temperature.

The ratio of the rates of reaction for the heavy and light isotopes deﬁnes
the fractionation factor, a. Fractionation factors for photosynthesis (H2180/H2‘60
= 1.000 t 0.003, Stevens et al. 1975, Guy at al. 1993), respiration (180-160/‘60-
‘60 = 0.9770, Luz et al. 2002), gas exchange (‘80-‘60/‘60-160 = 0.9770, Knox et
al. 1992) and gas dissolution (‘80-‘60/‘60-‘60 = 1.0007, Kroopnick and Craig,
1972; Benson and Krause, 1984) are indicated by up, err, a9, and ors respectively.
Respiration can occur via four pathways in aquatic ecosystems, Mehler reaction,
photorespiration, cytochrome oxidase, and alternative oxidase (Kroopnick and
Craig, 1972; Kroopnick ,1975; Guy et al. 1989; Kiddon, 1993). Therefore, orr is a
net respiratory fractionation factor representing the combined isotope effects of
the respiration pathways on dissolved 02. In the calculation for R:P, we used an
estimated orr value of 0.9770 determined by Luz et al. (2002) for epilimnetic
waters in Lake Kinneret as representing a phytoplankton assemblage and trophic

state similar to those of Lake Erie.

RESULTS

West to East Transect

Temperature proﬁles exhibited an overall cooling trend on a west to east
transect across the lake (Table 1, Figure 3). Temperatures were near 25°C
throughout the water column at station 91 M indicative of mixing in the Shallow
western basin, while thermal stratiﬁcation was present in the central and eastern
basins. At station 43 in the central basin a marked therrnocline was present at a
depth of 15 m where the water temperature was near 17°C. Epilimnetic waters
at this station were approximately 25°C while temperatures as low as 12°C were
evident in the hypolimnion. The temperature proﬁle at central basin station, 78M,
followed a similar trend with depth to that present at station 43. A deeper and
less deﬁned therrnocline was evident at a depth of 20 m in the eastern basin at
station 15M. Temperature ranged from 18°C to 25°C above the therrnocline while
the hypolimnion in the eastern basin was considerably cooler relative to the other
basins with bottom waters approaching 5°C.

A trend of decreasing chlorophyll-a ﬂuorescence (expressed in relative
ﬂuorescence units (RFU)) was apparent from west to east across the lake
(Table 1, Figure 3). RFU values were substantially higher in the western basin
than in the other two basins with values as high as 2.7 consistent with relatively
high productivity levels and nutrient loading (Charlton and Milne, 2005; Ostrom et
al. 2005a; b). In the central basin ﬂuorescence varied from 0.5 to 1.8. The
lowest RFU values were observed in the eastern basin with values between 0.04

and 1.2. Fluorescence increased with depth in the mixed water column at station

10

91 M, while at stations 43, 78M and 15M ﬂuorescence decreased moderately with
depth.

A decreasing trend in 02 saturation with depth was apparent at all stations
(Table 1, Figure 4). The variation in 02 concentration with depth at station 91 M
was minimal (0.7 to 1.1), which likely reﬂected extensive water column mixing at
this shallow station. At the central basin station 43 02 saturation (0.9 to 1.1)
remained at or near equilibrium with the atmosphere throughout most of the
epilimnion, however, intense undersaturation of O2 in the hypolimnion was
evident by values near 0.1. 02 saturation in the epilimnion at 78M (0.8 to 1.0)
was similar to that present at station 43, however, the degree of O2 depletion in
the hypolimnion was slightly less intense as indicated by values near 0.3. These
saturation values equate to approximately 33 and 92 pmollL at stations 43 and
78M respectively, indicating that hypoxia was present in the central basin at
station 43, while only near hypoxic conditions were observed further east at
station 78M. At station 15M in the eastern basin, 02 saturation values ranging
from 0.7 to 1.1 throughout the water column were more homogenous relative to
the central basin stations. 02 saturation values (>07) in the bottom waters at
station 15M were indicative of O2 consumption by respiration, however, the
values were much greater than those observed within the central basin
hypolimnion. The difference in volume of the hypolimnion and, therefore, total
mass of 02 within the hypolimnion, indicates that 02 depletion is more readily
accomplished within the central basin relative to the eastern basin (Charlton and

Milne, 2005). Lake-wide values within the epilimnion ranged between

I]

approximately 0.8 and 1.1, which indicates a strong inﬂuence of atmospheric
exchange on 02, while values in the hypolimnion were exclusively less then 1.0,
showing a strong inﬂuence of respiration.

A framework for understanding isotopic variation in dissolved 02 can be
gained based on the premise that a value of 0.7 %o reﬂects atmospheric
equilibrium and that biological processes cause deviation from this value (Bender
and Grande, 1987). The isotopic composition of 02 across the lake ranged from
-4.0 to 12.9 %o and in general an increasing trend in the 6180-02 was evident
throughout the water column at all stations (T able 1, Figure 4). Little Change in
isotopic composition with depth was apparent at 91 M (-1.5 and -2.7 960) which is
likely the result of mixing given the shallow nature of the water column at this
station. Values consistently less than 0.7 %o observed throughout the water
column indicate primary production is predominate in the western basin. At
station 43 isotope values were highly variable and ranged from -4.0 %0 near the
surface to 12.9 960 in deep water. A similar trend was found at station 78M with
one noticeable difference; there was a slight decrease in 8‘80 near the bottom of
the lake following a steady increase in 6180-02 values at shallower depths. This
isotope shift may be an artifact of collecting samples from a thin hypolimnion as
water from above the intended sampling depth is not thoroughly ﬂushed from the
Niskin bottle before it is triggered. At station 15M, 6180-02 values ranged from
-2.3 to 4.9 960. In the eastern basin the trend in 6180 increased with depth,
however, the range in isotope values below the thermocline was narrow relative

to the central basin. In general, negative isotope values (<0.7 960) observed near

12

the lake surface suggest high levels of primary production whereas the
increasing trend in 8180-02 indicates respiration is the controlling O2 cycling
process at depth in Lake Erie. Overall trends in 8180-02 exhibited considerable
variation with depth at each station as well as across the lake.

In this paper we only report R:P values for epilimnion samples as
hypolimnion values are not applicable to the evaluation of trophic state. R:P
ratios observed in the epilimnion of Lake Erie were very near 1.0 with a narrow
range from 1.0 to 1.2 (Table 2). This trend indicates that the rates of
photosynthesis and respiration were approximately equal across the lake

surface, with a slight predominance of net heterotrophy.

13

Diurnal Study

The concentration and isotopic composition of dissolved 02 was measured
at both central basin stations in Lake Erie over a diurnal cycle (Figure 5, 6). At
station 43, 02 saturation, 8180-02, and R:P were determined 6 times over a 24 h
period beginning at 9:15 pm. At station 78M, the same parameters were
determined 5 times over a 24 h period beginning at 1:15 am. The diurnal study
was conducted for the purpose of observing the inﬂuence of daytime and
nighttime light conditions on the metabolic balance in the central basin.

Values of 02 saturation at station 43 were as high as 1.18 in the
epilimnion and as low as 0.06 in the hypolimnion over the diurnal cycle (Figure
5). A decrease in 02 concentration was evident with depth for each sampling
time indicating a strong consumption of 02 by respiration. Slight O2
undersaturation was evident at the surface for each sampling event (0.76 to
0.98). This was followed by a peak in the [O2] proﬁle observed between 2 and
4 m, in which Slight supersaturation was evident (1.03 to 1.18) with one
exception; the maximum value observed for the 8:00 pm sampling event was
0.95. Average epilimnetic 02 saturation below 3 m (3 to 11 m) decreased
gradually between 9:15 pm and 9:05 am (1.03, 1.01, 0.96, and 0.91,
respectively). This was likely due to respiration in the upper water column at
night when light conditions do not support photosynthesis. Saturation values for
the epilimnion were predominately near 0.95 for both the 5:00 pm and 8:00 pm
sampling events. A predominance of values near 1 throughout the epilimnion

over the diurnal period may be the result of mixing by high winds associated with

14

the passing of a storm. Hence, marked variation in [02] concentration (and 8130)
may be the result of mixing by the storm or due to storm induced internal waves.
Hypolimnetic waters sampled at each time were Characterized by values less
than 0.09 indicating a predominance of respiration.

The 8180-02 values determined for epilimnion samples collected over the
diurnal cycle at station 43 were consistently less than 0.7 960, while values for the
hypolimnion were highly enriched in 180 (Figure 6). The lowest isotope value
determined for the epilimnion (4.0 %0) was observed at 9:15 pm 6130-02
generally increased between 9:15 pm and 5:00 pm from -4.0 to -1.5 %o, followed
by a decrease to a value of 3 %o observed at 8:00 pm. Observation of the lowest
isotope values in the early evening likely indicated a temporal decoupling
between elevated daytime primary production and the isotope signal. For each
sampling event over the 24 h period, 5180-02 increased to a depth of at least 16
m (near the thermocline) indicative of an increase in the predominance of
respiration over production with depth. The increase, however, was greater over
the nighttime sampling period between 9:15 pm and 9:05 am (16 m values
increased 2.7 960) than after a period of daytime production (16 m values
increased 0.8 %o from 5:00 and 8:00 pm). These values are reﬂective of greater
O2 consumption during the night when photosynthesis is absent. An anomalous
decrease (except at 1:05 am) in the isotopic composition of 02 occurred near the
bottom of the lake where the value was expected to be the highest. This

anomaly likely resulted from the tendency for the Niskin bottle to trap water from

15

multiple rather than discrete depths in the narrow and O2 deﬁcient hypolimnion of
the central basin.

At station 78M 02 saturation trends were similar to those observed at
station 43 and in general little variation was evident between sampling events
(Figure 5). In contrast to station 43, sampling at 78M occurred during a period of
calm winds. Concentrations of 02 near the surface were consistently
undersaturated for each time sampled. The highest degree of undersaturation
occurred at night with values of 0.78 and 0.67 observed at 1 m for the 1:15 am
and midnight sampling events. This trend likely reﬂects nighttime respiration in
the absence of photosynthesis. 02 saturation values determined for the
epilimnion (below 3 m) over the entire diurnal cycle were consistently near 1,
indicating the importance of atmospheric exchange. Near hypoxic conditions
were evident in the bottom waters as indicated by low saturation values between
0.2 and 0.3, however, the extent of O2 depletion was less than that observed at
station 43.

6180-02 values indicated photosynthesis was the primary process
controlling O2 in the epilimnion, while values for the hypolimnion suggested 02
was consumed by respiration at depth throughout the diurnal cycle (Figure 6).
Isotope values for the epilimnion over the 24 h period were -3.5 (1 :15 am), -2.8
(6:05 am), -2.7 (12:05 pm), -3.8 (6:05 pm), and -2.9 %o(12:00 am). The low
isotope value observed at 1:15 am indicates a lag in time for the isotope signal to
respond to daytime primary production. An increasing shift was evident by

morning as a result of nighttime respiration. The lowest isotope value was

16

observed at 6:05 pm likely following a period of high primary production
throughout the day. Trends similar to that observed at station 43 occurred in the
hypolimnion. Hypolimnion 6180-02 values for nighttime and early morning
sampling events (8.1 to 10.2 960) were higher than those observed during the day
(5.6 to 7.5 960) indicating a greater degree of respiration over night. Furthermore,
the same decreasing anomaly was evident near the bottom of the lake at station

78M.

17

DISCUSSION AND CONCLUSIONS

Hypoxia observed in the central basin of Lake Erie is largely driven by
respiration of organic matter produced during photosynthesis rather than that
derived from the watershed (Charlton et al. 1993; 1999; Charlton and Milne,
2005; Ostrom et al. 2005b). A study similar to this was conducted in Lake Erie in
2002, which Characterized July as a period of net autotrophy whereas net
heterotrophic conditions and the development of hypoxia were observed in
August (Ostrom et al. 2005b). This study focussed on O2 cycling and the
metabolic balance between production and respiration in Lake Erie in August
2003.

The variation in abundance and isotopic composition of 02 observed in
Lake Erie suggests a strong inﬂuence of biological activity on lake-metabolism
relative to gas exchange. The relationships between photosynthesis, respiration,
and atmospheric gas exchange and their inﬂuence on dissolved 02 in the lake
can be understood by use of a quad plot (Quay et al. 1995; Field, 2004; Ostrom
et al. 2005b) (Figure 7). Within this plot four quadrants are produced by the
intersection (deﬁned as the equilibrium locus) of lines representing 02 saturation
(1.0) and isotopic composition (0.7 %o) associated with atmospheric equilibrium.
These values likely represent conditions in the lake during winter and early spring
when low light levels and cold temperatures limit biological activity and physical
mixing favors equilibrium with the atmosphere (Russ et al. 2004, Ostrom et al.
2005b). The onset of primary production and respiration in the spring and

summer months shift the concentration and isotopic composition of 02 away from

18

the equilibrium locus. Predominance of photosynthesis over respiration will drive
values into quadrant IV, which is characterized by 02 saturation values greater
than one and isotope values less than 0.7 %o indicative of 02 production. When
consumption by respiration is the predominant process controlling O2
abundance, values will fall within quadrant II and be characterized by 02
saturation values less than 1.0 and 6180-02 values greater than 0.7 %o. The
magnitude of isotopic fractionation during respiration is large while little to no
fractionation is associated with photosynthesis during which 180 depleted O2 is
produced. Consequently, if photosynthesis and respiration occur at the same
rate a net isotope shift will occur even though concentration remains constant.

Epilimnion samples obtained for this study lie within or near quadrant IV of
the quad plot reﬂecting a predominance of photosynthesis at the surface while
hypolimnion samples lie within quadrant ll indicating a strong inﬂuence of
respiration at depth. The epilimnion values are unique from those common in the
oligotrophic Lake Superior and Grand Traverse Bay in which values reﬂective of
atmospheric exchange were predominant (Field, 2004; Russ et al. 2004).
Consequently, the wider range of 8180 values and frequency of negative values
in the epilimnion is consistent with a eutrophic ecosystem. Furthermore, the
occurrence of values in quadrant II and Ill reﬂect the important inﬂuence
respiration has on O2 cycling in Lake Erie particularly within the hypoxic waters of
the central basin.

Metalimnion samples transition between quadrant II and IV predominately

occurring in quadrant lll. If the equilibrium locus represents the concentration

and isotopic composition of dissolved 02 prior to initiation of biological activity in
the spring then values should not occur in quadrants l or III. While no values
were observed in quadrant l, metalimnion values did in fact, lie within quadrant
III. This trend suggests that the concentration and isotope values associated
with atmospheric equilibrium do not deﬁne the equilibrium locus of the Lake Erie
system much of the time. The locus in this system is instead shifted towards
negative isotope values, which suggests an inﬂuence of photosynthesis on the
“locus” values for Lake Erie. A shift in locus values may reﬂect the initiation of
primary production in Lake Erie prior to thermal stratiﬁcation in the spring
(Ostrom et al. 2005b). The occurrence of values in quadrant III of the quad plot
and thus a shift in the locus from atmospheric equilibrium was also observed in
Lake Erie during the summer of 2002 (Ostrom et al. 2005b). Furthermore, a
similar trend was evident in Lake Superior and Grand Traverse Bay with the
exception that isotope values deﬁning the locus reﬂected a predominance of
respiration (Field, 2004; Russ et al. 2004). The distinction in locus values for
Lake Erie relative to the more oligotrophic systems, Lake Superior and Grand
Traverse Bay, is consistent with its eutrophic nature.

If the hypolimnion can be considered a closed system once stratiﬁcation is
established the Rayleigh model can be used to determine the fractionation factor
for respiration (Ostrom et al. 2002; 2005b). The lack of 02 production in the
hypolimnion supports the assumption that Lake Erie is indeed a closed system
(Ostrom et al. 2005a). Furthermore, diffusion of 02 across the thermocline is a

slow process and, therefore, is not likely a signiﬁcant source of O2 to

20

hypolimnetic waters (Emerson et al. 1999); particularly in a system Characterized
by high rates of photosynthesis and respiration (Carrick, 2004; Ostrom et al.
2005a). In a closed system in which only one process is controlling the isotopic
signature of 02, the fractionation factor, 8, can be calculated using a modiﬁed

Rayleigh equation,

(3) 63 = 55° - SII'I(C/Co)

where s is deﬁned as (1/or-1)*1000, or is the reaction rate for 02 containing the
light and heavy isotopes, respectively, 83 and 83° represent the isotopic
composition of the residual and initial substrate of a reaction and C/Co is the ratio
of the observed to initial substrate concentration, respectively (Mariotti et al.
1981; Ostrom et al. 2002; 2005b). C/Co can be replaced by the 02 saturation
value based on the assumption that the initial concentration reﬂects atmospheric

equilibrium at the hypolimnetic temperature. Thus equation 3 becomes:

(4) as = as, - s|n(O2 Sat.)

If 53 is plotted versus -ln(O2 Sat.) then the slope of the line deﬁnes the

fractionation factor, 8, associated with respiration (Ostrom et al. 2002; 2005b).
The slopes of hypolimnion data plotted for stations 78M and 15M yielded

fractionation factors of 9.0 and 8.7 960, respectively. Owing to a poor correlation

among values (R2 = 0.1) s from station 43 data was omitted. A poor correlation

21

for station 43 data may reﬂect the Challenge of measuring 8180 at low 02
concentrations as well as the asymptotic nature of the Rayleigh model
relationship at low 02. The fractionation factors obtained are very similar to the
preliminary values based on only a few data points reported by Ostrom et al.
(2005b). The value of approximately 9.0 %o obtained for hypolimnion respiration
is markedly different than the value of 23.5 960 reported by Luz et al. (2002). A
fractionation factor for sediment respiration of 3.0 960 was reported by Brandes
and Devol (1997). This low a value is the result of diffusion limiting the supply of
O2 to the heterotrophic community in sediments. The low 8 value of 9.0 %o in
Lake Erie, therefore, indicates a strong inﬂuence of sediment respiration on
hypolimnetic 02. Based on a values of 23.5 %o for water column respiration and
3.0 %o for sediment respiration we calculate that 71 % of hypolimnetic respiration
occurs in sediments. This indicates that the development of hypoxic conditions
in the central basin of Lake Erie is largely driven by sediment respiration. The
predominance of sediment respiration in controlling the development of hypoxia
within the central basin may, in part explain why the deeper eastern basin is not
susceptible to hypoxia.

The R:P ratio correlates with total phosphorus and chlorophyll-a
concentrations in a variety of lakes and has therefore been proposed as a trophic
state indicator (Jones, 1992; del Georgio and Peters, 1993;1994). Considering
primary production in Lake Erie is largely restricted to the epilimnion we report
R:P values for only the upper water column (Ostrom et al. 2005a). R:P for the

epilimnion of Lake Erie in August 2003 ranged from 1.0 to 1.2 (Table 2). These

22

values indicate that in Lake Erie primary production and respiration are in
balance with only a slight predominance of net heterotrophy. A period of net
heterotrophy in August was also observed by Ostrom et al. (2005b) in 2002,
following a period of net autotrophy in July. Furthermore, a similar oscillation
between periods of net autotrophy and heterotropy was observed in the
oligotrophic environments, Lake Superior and Grand Traverse Bay (Field, 2004;
Russ et al. 2004, Ostrom et al. 2005b).

The seasonal variation in lake metabolism observed in Lake Erie, Lake
Superior and Grand Traverse Bay indicates that R:P may not provide an
accurate measure of trophic state. Respiration in oligotrophic systems is strongly
affected by terrestrial inputs of organic carbon, and thus these environments tend
to be net heterotrophic. Lake Erie is well recognized as a eutrophic system and
is, therefore, expected to be net autotrophic. The R:P values we determined,
however, are clearly not consistent with a net autotrophic system. The
observation of net heterotrophy in a eutrophic system can be best explained due
to temporal and spatial decoupling of photosynthesis and respiration. For
example, organic matter produced in the spring may be respired at a later time or
sediment resuspension events may further the respiration of material produced
years to decades ago (Field, 2004; Russ et al. 2004, Ostrom et al. 2005b). The
presence of high rates of production and algal blooms observed in the western
basin of Lake Erie combined with the potential for transport of material during
storm events in this relatively small and shallow lake may be a mechanism

whereby material produced in one portion of the lake supports respiration in

23

another (Carrick et al. 2005; Ostrom et al. 2005a; b).

The wide range in 6180-02 and [02] observed in Lake Erie indicates a
predominance of biological activity relative to inputs via atmospheric exchange.
A much narrower range in 6180-02 and [02] values was evident in Lake Superior
and Grand Traverse Bay (Figure 8). Within these systems 02 cycling was
controlled predominately by gas exchange (Field, 2004; Russ et al. 2004). A
better indicator of trophic state may be the relative importance of gas exchange
versus biological activity. Thus the wide range of 8180-02 and [02] values
evident in Lake Erie relative to the oligotrophic lakes Lake Superior and Grand

Traverse Bay deﬁne the lake as a eutrophic system.

24

Table 1: Data summary for samples collected in August, 2003.

 

Sample 02

 

Date Station Depth Time Tsmp O2 Concen 5 180
(2003) Number (m) (EST) ( C) Saturation (umol IL) (960, wrt aIr)
15—Aug 91 M 1 0:15 25.6 0.98 260.2 -2.7
15-Aug 91M 8 0:10 24.2 0.81 219.5 -1.5
16-Aug 15M 1 5:00 24.6 1.01 272.8 -2.3
16-Aug 15M 10 5:00 21.2 0.89 250.8 0.1
16-Aug 15M 28 5:00 6.8 0.82 313.5 3.6
16—Aug 15M 40 5:00 5.2 0.76 304.1 4.2
16-Aug 15M 61 5:00 5.1 0.71 285.3 4.9
16-Aug 43 5 21:15 25.0 0.87 232.0 -4.0
16-Aug 43 12 21:15 22.1 0.31 87.8 4.3
16-Aug 43 16 21:15 12.2 0.05 15.7 9.6
16-Aug 43 20 21:15 11.8 0.07 25.1 4.2
17-Aug 43 5 1:05 24.7 1.00 268.7 -3.0
17-Aug 43 12 1:05 23.8 0.57 154.2 -0.9
17-Aug 43 16 1:05 12.3 0.10 34.5 10.3
17-Aug 43 20 1 :05 11.9 0.07 22.6 12.9
17-Aug 43 5 5:05 24.4 0.83 223.5 -1.5
17-Aug 43 12 5:05 24.0 0.32 86.2 nd
17-Aug 43 16 5:05 11.9 0.04 14.4 12.3
17-Aug 43 20 5:05 11.9 0.05 17.9 6.8
17-Aug 43 5 9:05 23.7 0.82 224.5 -1.5
17-Aug 43 12 9:05 22.3 0.60 165.8 -0.9
17-Aug 43 16 9:05 11.6 0.08 26.0 11.4
17-Aug 43 20 9:05 11.6 0.07 22.6 4.5
17-Aug 43 5 13:05 24.3 0.78 211.9 -2.6
17-Aug 43 12 13:05 23.4 0.21 57.4 2.7
17-Aug 43 16 13:05 11.7 0.06 21.6 0.8
17-Aug 43 20 13:05 11.6 0.07 25.1 9.5
17-Aug 43 5 17:00 24.5 0.89 239.5 -1.5
17-Aug 43 12 17:00 23.0 0.46 126.3 0.0
17-Aug 43 16 17:00 11.7 0.11 36.4 7.2
17-Aug 43 20 17:00 11.6 0.10 32.9 3.8
17-Aug 43 5 20:00 24.4 0.88 237.0 -3.1
17-Aug 43 12 20:00 23.5 0.20 53.9 -3.2
17-Aug 43 16 20:00 11.9 0.06 20.4 8.0

17-Aug 43 20 20:00 11.7 0.11 38.9 3.4

25

 

 

Sgrgge Station Depth Time Timp O2 Cori; n 5 ‘30
(2003) Number (m) (EST) ( C) SaturatIon (pmol IL) (960, wrt aIr)
18—Aug 78M 5 1:15 23.9 1.03 280.3 -3.5
18-Aug 78M 15 1:15 16.0 0.28 85.6 9.0
18-Aug 78M 18 1:15 12.7 0.25 81.8 9.4
18-Aug 78M 21 1:15 12.7 0.25 83.7 8.1
18-Aug 78M 5 6:00 23.8 0.94 256.8 -2.8
18-Aug 78M 15 6:00 12.8 0.25 81.5 9.5
18-Aug 78M 18 6:00 12.7 0.25 82.1 8.5
18-Aug 78M 20 6:00 12.8 0.28 93.1 9.3
18-Aug 78M 5 12:05 23.8 0.95 259.3 -2.7
18—Aug 78M 15 12:05 20.0 0.33 96.3 6.4
18-Aug 78M 18 12:05 12.7 0.27 90.9 nd
18-Aug 78M 20 12:05 12.6 0.36 120.7 nd
18-Aug 78M 5 18:05 24.7 1.07 286.9 -3.8
18-Aug 78M 15 18:05 17.9 0.34 100.6 6.1
18-Aug 78M 18 18:05 12.8 0.29 94.7 7.5
18-Aug 78M 20 18:05 12.7 0.35 116.0 5.6
19-Aug 78M 5 0:05 23.8 0.97 264.0 -2.9
19-Aug 78M 15 0:05 16.6 0.32 97.2 nd
19-Aug 78M 18 0:05 12.7 0.23 74.6 10.2

19-Aug 78M 20 0:05 12.7 0.23 75.2 9.7

26

Table 2: R:P ratios calculated for the Lake Erie epilimnion.

 

Sample Date Station Depth Time RP
(2003) Number (m) (EST) '

 

15-Aug 91M 1 0:15 1.0
15-Aug 91M 8 0:10 1.1
16-Aug 15M 1 5:00 1.0
16-Aug 15M 10 5:00 1.2
16—Aug 43M 5 21:15 1.0
17-Aug 43M 5 1 :05 1.0
17-Aug 43M 5 5:05 1.1
17-Aug 43M 5 9:05 1.1
17-Aug 43M 5 13:05 1.1
17-Aug 43M 5 17:00 1.1
17-Aug 43M 5 20:00 1.1
18-Aug 78M 5 1 :15 1.0
18-Aug 78M 5 6:00 1.0
18—Aug 78M 5 12:05 1.0
18-Aug 78M 5 18:05 1.0
19-Aug 78M 5 0:05 1.0

 

27

Base 8 ”BOOB as .255 as .23

new Esme mm iron. 5 .258 NS 2R Esme Sr. ”2,058 5 .2024 E 9 Ease
m <5“.on mm .zovdm _.$ .25 M226“— mm 2m :03ma some Co EEOC 63:..me new 865208
on... .etw 9.3 c. Cowman: $3.50? a 9:85— mcosﬂm 959:8 Co 20:80.. ”P 059m

 

z and:

z .cooNv

z .cnoNV

 

 

 

_ _ _ _
>> began >> .090» 3 beer» 3 .2.on >> beamw

 

28

 

 

1.2

:5 1.0

E

E 0.8

8 0.6

3;

t!

3 0.4

N

(D

a 0.2
0.0

 

 

o
_ <>
9 0 o
- <><>
_. . I
I
‘ A
‘ I. ll
_ a - ,
at: A
0.0 0.2 0.4 0.6 0.8 1.0 1.2
02 Saturation (CTD)

 

Figure 2: 02 saturation determined from CTD measurements plotted
versus 02 saturation determined from Winkler titrations. The line

represents a 1:1 relationship. Open diamonds are epilimnion samples for
all stations, grey squares are metalimnion samples from stations 43 and
78M, solid triangels are hypolimnion samples from stations 43, 78M, and

15M.

29

 

dc: 8:22 m 3 “32:39am: m_ cozﬁm some um Egon E:E_me on... .95 :2 5v Cowman:

 

53.603 m 9.2m cam—u Co cozoce m we Sumv 85820:: m-__>;aodo_co new GOV 0.29an0... ”m 939“.

 

 

 

 

I a mo
00
mm
on

    

I

lull traded

I
It)
FFNNOOOOVV

I

I
OIOOLOOIDOIOO

 

 

W q _

m N _. o
5%: 0283202....

 

 

NN
. 0N
. 9.
r 0..
r 3.

 

a
(w) Indoo

I

I
ONV’CDCDO
‘—

 

 

_ q a

m N F o
SHE. 3:33.32...

 

 

aNN
row
IS.

I
(D
‘—

le undoo

I

I

T

I
ONVOQONV‘

 

 

_ u a

m N P 0
5mm. 3:33.02“.

 

nNN
ION
-9.
r9.
r3.
IN?
Tow

 

 

— u _

m N r c
any: 3:33.02“.

(w) uzdoo

 

 

 

 

 

 

 

 

 

 

 

 

 

1 — ~

on ON or o
.0... 2322.50...

 

 

mm
7 cm
r 3
r 9.
r S.
r N?

oasazé
(w) urdoo

1

 

 

 

1 _ A

on ON or o
.0... 23anth

 

 

INN
row
.9.
-9.
r3
.9.

 

(w) llldOO

I

I

I
ONV'CDQO
‘—

 

 

_I q _

om ON 0.. 0
Ga. 2382.83

 

-NN
row
r9.
r9.
r3.
rNF

I

 

I

I

I
ONVOQO
‘—

 

 

_ A ~

on ON 0.. o
8... 2322.95...

 

(“1) H1690

 

 

 

 

 

Esnem . :2 538m

.3230 . Emu cowﬂm

35:00 . M? 20395

£382, - :3 sees..."

30

.05. U055:

m B beacon-OED. m_ 226% some um Egon 25.5me of. .95 75:3 Semen: ammo-“mo; m 9.2m canon Co c0305.:
m mm NO-OEw cam 20:: >89 9.0 new Amocmacmv c0595 .m_v_c_>> E umEmEo ENE con—82mm NO “v 2:9".

 

 

 

 

 

amo
-00
-mm

I
0
ID

I I I I

I

I

rvNNmmvv

 

 

T‘r I I
OlDOlDOl-DOIOOIO

— H _ Idl

me o_. m o m-
7.3:...

(w) "1690

 

 

 

 

 

 

 

-NN
-0N
-m_.
-o_.
-v_.
-N_.
-ow

I

(w) urdoo

I

I

r
ONVCOQ

 

 

— q a d

2 2 m o m-
3...?»

 

-NN
-0N
-m_.
-o_.
-v_.

I
N
F

I

(w) llldOCl

 

I

I

I
ONV’CDCDO
‘—

 

 

Bosses-
BLPR

 

 

 

 

 

 

 

 

 

 

I I I I I I I 7
IOOIDOID
IDIOOCD

I

I

I
OIOOIOOIOOIOO
PFNNC‘Ot‘OVﬁ'

I

I

 

 

N... md v.0 od
cow—233 No

 

(w) Indoo

 

 

 

NN
-o~
-w_.
-o_.
arm
-N_.N
.ovw
le‘
To
-v
rN
q q a a o
9. or m o m-
029%
mm
-ON
-m_.
-©—.a
-v_.®
-N_.m.e
-9.)
NM
-0
-v
-N
. . . o

 

 

Ne we to Re
ease—=86

 

 

-NN

I I
coo
s-N

I I I

oa.aaéggg
(w) Indoo

I

I

 

 

3 we we 90
532386

 

 

 

 

Ne wd v.0 o.o
cozﬂaﬁm No

 

 

 

 

 

£38m - :2 8:3

.9530 - Emu town—m

.828 - 9 5.5.0-

5883 - :3 5:8

3I

 

02 Saturation

 

 

 

 

time (hours)
[—0-5m +12m +16m +20m]

 

 

 

02 Saturation

 

 

time (hours)

 

+5m -E-15m -ﬁ-18m -0-20m

 

 

 

Figure 5: 02 saturation plotted versus time for samples collected over a 24-
hour diurnal cycle at central basin stations 43 and 78M. The shaded area
represents nighttime conditions. Samples were collected at
5m(diamonds), 12m(squares), 16m(triangles), and 20m(circles) at station
43 and 5m(diamonds), 15m(squares), 18m(triangles), and 20m(circles) at
station 78M.

32

 

43 14 - ..

5180-02 (°I,,,,)
A

 

.L
4

time (hours)

 

+5m +16m +20m +12m

 

 

78M

 

51°04), I°I..)

 

 

 

 

time (hours)

 

 

-0-5m -E-15m -¢-18m +20m

 

 

Figure 6: 6180-02 plotted versus time for samples collected over a
24-hour diurnal cycle at central basin stations 43 and 78M. The shaded
area represents nighttime conditions. Samples were collected at
5m(diamonds), 12m(squares), 16m(triangles), and 20m(circles) at station
43 and 5m(diamonds), 15m(squares), 18m(triangles), and 20m(circles) at

station 78M.

33

 

 

12 - a” y = -8.9Ln(x) - 3.2
R2 :10
e

  
   
    

y = ~21.4Ln(x) - 1.9
R2 = 0.9

5"’o-o2

 

 

 

y = -3.9Ln(x) - 3.1
R2 = 0.7
'8 III I l I I I I
0.0 0.2 0.4 0.6 0.3 1.0 1.2

02 Saturation

 

 

 

Figure 7: 8180-02 plotted versus 02 saturation for stations 91 M(circles),
43(open squares), 78M(open diamonds), and 15M(triangles).

34

 

 

 

 

 

 

 

A
12 . A,
‘2‘ ‘t‘
_ A
A A D
C? 6 ~ A As
4 - I
O “ A
:2 2‘ a I
'0 Old A
AA
-2 .
_4 ﬂ A
'6 _ III
’8 f f I
0.0 0.2 0.4 0.6

 

02 Saturation

 

 

Figure 8: 6180-02 plotted versus 02 saturation reported for Lake Erie 2003

(triangles), Lake Eire 2002 (squares) (Ostrom et al., 2005), Grand Traverse
Bay (open diamonds) (Field, 2004), and Lake Superior (circles) (Russ et
aL,2004)

35

REFERENCES

Bender, ML. and Grande, K.D. 1987. Production, respiration, and the isotope
geochemistry of O2 in the upper water column. Global Biogeochem.
Cycles 1: 49-59.

Benson, BB. and Krause, D. 1984. The concentration and isotopic fractionation
of oxygen dissolved in freshwater and seawater in equilibrium with the
atmosphere. Limnol. Oceanogr. 29: 620-632.

Bertram, PE. 1993. Total phosphorus and dissolved oxygen trends in the
central basin of Lake Erie, 1970-1991. J. Great Lakes Res. 19: 224-236.

Biddanda, BA. and Cotner, J.B. 2002. Love handles in aquatic ecosystems:
The role of dissolved organic carbon drawdown, resuspended sediments,
and terrigenous inputs in the carbon balance of Lake Michigan.
Ecosystems. 5: 431445.

Brandes, J.A. and Devol, AH. 1997. Isotopic fractionation of oxygen and
nitrogen in coastal marine sediments. Geochim. Cosmochim. Acta. 61:
1793-1801.

Carignan, R., Planas, D. and Vis, C. 2000. Planktonic production and
respiration in oligotrophic shield lakes. Limnol. Oceanogr. 45: 189-199.

Carpenter, J.H. 1965. The accuracy of the Winkler method for dissolved oxygen
analysis. Limnol. Oceanogr. 10: 135-143.

Carrick, H.J. 2004. Algal distribution patterns in Lake Erie: Implications for
oxygen balances in the eastern basin. J. Great Lakes Res. 30: 133-147.

Carrick, H.J., Moon,J.B. and Gaylord, B.F. 2005. Phytoplankton dynamics and
hypoxia in Lake Erie: evidence for benthic—pelagic coupling in the central
basin. J. Great Lakes Res. 31(Suppl. 2): 111-124.

Charlton, M.N. 1980a. Oxygen depletion in Lake Erie: Has there been any
change? Can. J. Fish. Aquatic Sci. 37: 72-81.

Charlton, M.N. 1980b. Hypolimnion oxygen consumption in lakes: discussion of
productivity and morphometry effects. Can. J. Fish. Aquatic Sci. 37:
1 531 -1 539.

Charlton, MN. and Lean, D.R.S. 1987. Sedimentation, resuspension, and
oxygen depletion in Lake Erie (1979). J. Great Lakes Res. 13: 709-723.

36

Charlton, MN. 1987. Lake Erie oxygen revisited. J. Great Lakes Res. 13: 697-
708.

Charlton, M.N., Milne, J.E., Booth, WC. and Chiocchio, F. 1993. Lake Erie
offshore in 1990: Restoration and resilience in the central basin. J. Great
Lakes Res. 19: 291-309.

Charlton, MN. and Milne, J.E. 2005. Review of thirty years of change in Lake
Erie water quality. NWRI contribution # 04-167. National Water Research
Institute, Environment Canada.

Cole, J.J., Pace, M.L., Carpenter, SR. and Kitchell, JP. 2000. Persistence of
net heterotrophy in lakes during nutrient addition and food web
manipulations. Limnol. Oceanogr. 45: 1718-1730.

del Giorgio, PA. and Peters, RH. 1993. Balance between phytoplankton
production and planktonic respiration in lakes. Can. J. Fish. Aquatic Sci.
50: 282-289.

del Giorgio, PA. and Peters, RH. 1994. Patterns in planktonic P:R ratios in
lakes: inﬂuences of lake trophy and dissolved organic carbon. Limnol.
Oceanog. 39: 772-787.

Emerson, S., Quay, P., Stump, C., Wilbur, D. and Knox, M. 1991. 02, Ar, N2,
and 22an in waters of the subarctic ocean: net biological 02 production.
Global Biogeochem. Cycles. 5: 49-69.

Emerson, S., Stump, C., Wilbur, D. and Quay, P. 1999. Accurate measurement
of 02, N2, and Ar gases in water and the solubility of N2. Mar. Chem. 64:
337-347.

Field, AL 2004. Seasonal variation in ratios of community respiration to gross
photosynthesis determined by stable isotopes and concentrations of
dissolved oxygen in Grand Traverse Bay, Lake Michigan. M. Sc. Thesis,
Department of Geological Sciences, Michigan State University.

Garcia, HE. and Gordon, Li. 1992. Oxygen solubility in seawater: better ﬁtting
equations. Limnol. Oceanogr. 37: 1307-1312.

Glooschenko, W.A., Moore, J.E., Munawar, M. and Vollenweider, RA. 1974.
Primary production in Lakes Ontario and Erie: a comparative study. J.
Fish. Res. Board Can. 31: 253-263.

Guy, R.D., Fogel, ML. and Berry, J.A. 1993. Photosynthetic fractionation of the
stable isotopes of oxygen and carbon. Plant Physiol. 101: 37-47.

37

Hanson, P.C., Bade, D.L. and Carpenter, S.R., Kratz, T.K. 2003. Lake
metabolism: relationships with dissolved organic carbon and phosphorus.
Limnol. Oceanogr. 48: 1112-1119.

Howarth, R.W. and Michaels, AF. 2000. The measurement of primary
production in aquatic ecosystems. In: Sala, 0.E., Jackson, R.B., Mooney,
HA. and Howarth, R.W. (eds) Methods in Ecosystem Science, Springer-
Verlag, New York, New York, pp. 72-85.

Jones, RI 1992. The inﬂuence of humic substances on lacustrine planktonic
food chains. Hydrobiologia. 229: 73-91.

Kiddon J., Bender M.L., Orchardo J., Caron D.A., Goldman J.C., Dennett M.
1993. Isotopic fractionation of oxygen by respiring marine organisms.
Global Biogeochem. Cycles. 7: 679-694.

Knox, M., Quay, PD. and Wilbur, D. 1992. Kinetic isotopic fractionation during
air-water gas transfer of 02, N2, CH4, and H2. J. Geophys. Res. 97:
20335-20343.

Kroopnick, PM. 1975. Respiration, photosynthesis, and oxygen isotope
fractionation in oceanic surface water. Limnol. Oceanogr. 20: 988-992.

Kroopnick, P. and Craig, H. 1972. Atmospheric oxygen: isotopic composition
and solubility fractionation. Science. 175: 54-55.

Lane, GA. and Dole, M. 1956. Fractionation of oxygen isotopes during
respiration. Science. 123: 574-576.

Laws, E.A., Landry, M.R., Barber, R.T., Campbell, L., Dickson, ML. and Marra,
J. 2000. Carbon cycling in primary production bottle incubations:
inferences from grazing experiments and photosynthetic studies using 1“C
and 18O in the Arabian Sea. Deep-Sea Res. ll. 47: 1339-1352.

Luz, B., Barkan, E., thach, S. and Yacobi, Y.Z. 2002. Evaluation of community
respiratory mechanisms with oxygen isotopes: a case study in Lake
Kinneret. Limnol. Oceanogr. 47: 33-42.

Makarewicz, J.C. and Bertram, P. 1991. Evidence for the restoration of the Lake
Erie ecosystem. Bioscience. 41: 216-223.

Mariotti, A., Germon, J.C., Hubert, P., Kaiser, P., Letolle, R., Tardieux, A. and
Tardieux, P. 1981. Experimental determination of nitrogen kinetic isotope
fractionation: some principles; illustration for the denitriﬁcation and
nitriﬁcation processes. Plant and Soil. 62: 413-430.

38

Odum, H. 1956. Primary production in ﬂowing waters. Limnol. Oceanog. 1:
102-117.

Ostrom, N.E., Hedin, L.O., von Fisher, J.C. and Robertson, GP. 2002. Nitrogen
transformations and nitrate removal at a soil-stream interface: a stable
isotope approach. Ecol. Appl. 12: 1027-1043.

Ostrom, N.E., Carrick, H.J., Twiss, MR. and Piwinski, L. 2005a. Evaluation of
primary production in Lake Erie by multiple proxies. Oceaologia. 144:
1 15-124.

Ostrom, N.E., Russ, M.E., Field, A.L., Piwinski, L., Twiss, MR. and Carrick, H.J.
2005b. Primary production and respiration in Lake Erie based on oxygen
isotope techniques. J. Great Lakes Res. 31(Suppl. 2): 138-153.

Prairie, Y.T., Bird, BF. and Cole, J.J. 2002. The summer metabolic balance in
the epilimnion of southeastern Quebec lakes. Limnol. Oceanogr. 47: 316-
321.

Quay, P.D., Emerson, s., Wilbur, DO. and Stump, C. 1993. The 5180 of
dissolved O2 in the surface waters of the subarctic Paciﬁc: a tracer of
biological productivity. J. Geophys. Res. 98: 8447-8458.

Quay, P.D., Wilbur, D.O., Richey, J.E., Devol, A.H., Benner, R. and Forsber, BR.
1995. The 180:160 of dissolved oxygen in rivers and lakes in the Amazon
Basin: determining the ratio of respiration to photosynthesis rates in
freshwater. Limnol. Oceanogr. 40: 718-729.

Roberts, B.J., Russ, ME. and Ostrom, NE. 2000. Rapid and precise
determination of the 6180 of dissolved and gaseous dioxygen via gas

chromatography-isotope rate mass spectrometry. Environ. Sci. Technol.
34: 2337-2341.

Rosa, F. and Burns, NM. 1987. Lake Erie central basin oxygen depletion
changes from 1929-1980. J. Great Lakes Res. 13: 684-696.

Russ, M.E., Ostrom, N.E., Gandhi, H. Ostrom, PH. and Urban, NR. 2004.
Temporal and spatial variations in R:P ratios in Lake Superior an
oligotrophic freshwater environment. J. Geophysical Research-Oceans.
109, doi: 10.1029/2003JC001890.

Stevens, C.L.R., Schultz, D., Van Baalen, C. and Parker, PL. 1975. Oxygen

isotope fractionation during photosynthesis in a blue-green and green
alga. Plant Physiology. 56: 126-129.

39

 

Sweeney, RA. 1993. “Dead” Sea of North America? - Lake Erie in the 19605
and ‘705. J. Great Lakes Res. 19: 198-199.

Verduin, J. 1956. Primary production in lakes. Limnol. Oceanog. 1: 85-91.

Wang, X. and Veizer, J. 2000. Respiration-photosynthesis balance of terrestrial
aquatic ecosystems, Ottawa area, Canada. Geochim. Cosmochim. Acta.
64: 3775-3786.

Weiss, RF. 1970. The solubility of nitrogen, oxygen and argon in water and
seawater. Deep-Sea Res. 17: 721-735.

40

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII