    

    

.7... 7V... .... M. . . n.

.v .. 74.. . H ...... . ........ .. ., .7 . .. I.qm‘l>h’o‘lﬁlR>tltl‘~4-‘

tzﬁ; llllHIHIHIJIZIILIIIHUIIHIUIUHHHllllllilllllllllllll

93 01812 9431

LIBRARY
Michigan State
Universlty

 

 

 

 

This is to certify that the

dissertation entitled

COMMUNITY METABOLISM IN MICHIGAN RIVERS: EFFECTS
OF INCLUDING HYPORHEIC SEDIMENTS IN ESTIMATES
OF PRODUCTION/RESPIRATION (P/R) RATIOS

presented by

Donald G. Uzarski

has been accepted towards fulﬁllment
of the requirements for

211.12. degreeinﬂﬁhﬁliﬁiand Wildlife

Major professor

DateW/ﬁ 77

MSU is an Afﬁrmative Action/Equal Opportunity Institution 0- 12771

VI... 1"

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

 

DATE DUE I DATE DUE

DATE DUE

 

NOV g9 {2009! a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/” WWW“

 

COMMUNITY METABOLISM IN MICHIGAN RIVERS: EFFECTS
OF INCLUDING HYPORHEIC SEDIMENTS IN ESTIMATES
OF PRODUCTION/RESPIRATION (P/R) RATIOS
By

Donald G. Uzarski

A DISSERTATION
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Fisheries and Wildlife

1999

ABSTRACT
COMMUNITY METABOLISM IN MICHIGAN RIVERS: EFFECTS
OF INCLUDING HYPORHEIC SEDIMENTS IN ESTIMATES
OF PRODUCTION/RESPIRATION (P/R) RATIOS
By

Donald G. Uzarski

The hypothesis tested in this study was that inclusion of a small portion of the
hyporheic zone could signiﬁcantly alter production/respiration (P/R) values, possibly to
the extent that some middle-order stream systems once believed to be autotrophic are
actually heterotrophic. I designed an open-ended hyporheic community metabolism
chamber and simultaneously measured metabolism using these hyporheic chambers and
standard in-situ benthic chambers. The standard benthic chambers were similar to those
used to develop the hypothesis of the river continuum concept (RCC) stating that P/R
values of medium size rivers are expected to be greater than one, indicating autotrophy.
The 100 cm long hyporheic chamber allowed water to ﬂow freely through the top 20 cm
of benthic sediments. A sediment transfer Shovel, specially designed to maintain
stratigraphic integrity was used to transfer sediments from the streambed to the chamber.
The chambers were left in place for 30 days for recolonization after sediment transfer.
Comparisons were made between the two chamber designs through a series of side by
side comparisons. Results from 4th and Sm-order Michigan rivers demonstrated that P/R
values using the traditional closed benthic chamber were always greater than one, as
predicted by the River Continuum Concept, while results obtained using the hyporheic

chamber were always less than one. These results, including a conservative 20 cm of the

hyporheic zone, lead to questions concerning P/R ratio changes along a river continuum.
Further, this experiment emphasizes the need to include the hyporheic zone in studies of
stream metabolism.

In a second set of studies, closed benthic chambers, whole system upstream-
downstream measurements, and the hyporheic chambers were tested simultaneously.
Several studies that had compared closed benthic chambers and whole system upstream-
downstream methods had suggested conﬂicting results. However, statistical comparisons
were not possible because the whole system method could not be replicated. The
hyporheic chamber was used along with both the upstream-downstream and closed
benthic chamber methods simultaneously at two sites on Augusta Creek, a third order
Michigan stream. The hyporheic chamber and upstream-downstream methods estimated
community respiration values higher than benthic chambers at both study sites.
Hyporheic and benthic chamber estimates were Signiﬁcantly different. Gross community
productivity was non-detectable using the hyporheic chamber at one site and was
signiﬁcantly higher than benthic chamber estimates at the other. The upstream-
downstream estimates of gross community productivity tended to be higher than benthic
chamber productivity estimates at both sites but tended to agree with hyporheic chamber
estimates. These data suggest inclusion of hyporheic sediments strongly inﬂuences
respiration estimates and indicate that whole system and benthic chamber methods should

not be use interchangeably.

ACKNOWLEDGMENTS

I would like to thank Dr.’S Thomas Burton, Donald Hall, Stephan Hamilton,
Daniel Hayes, and Robert King for input on the development of the hyporheic chamber
protocol. Dr. Stephan Hamilton also provided technical support, ﬁeld assistance, and the
use of his lab and equipment during these studies. I would like to thank Dr. Donna King
for the use of the closed benthic chambers, and Craig Stricker, Ted Siler, Mark
Scalabrino, Keith Kintigh, Christy Lewis, Donna Kashian, and Archie Martell for
assistance in the ﬁeld. Electra otec constructed and donated the hyporheic chambers.
Funding came from The Kalamazoo Foundation, Kalamazoo Valley Trout Unlimited, and

the US. Department of Defense.

TABLE OF CONTENTS

LIST OF TABLES ..................................................... vi
LIST OF FIGURES .................................................... vii
CHAPTER 1

INCLUDING 20 CM OF HYPORHEIC RESPIRATION

IN ESTIMATES OF PRODUCTION TO RESPIRATION ........................ 1
Introduction ...................................................... 1
Materials and Methods .............................................. 4
Results ......................................................... 11
Discussion ...................................................... 18

CHAPTER 2

METABOLISM IN A 3'd-ORDER STREAM:

A COMPARISON OF METHODS ......................................... 21
Introduction ..................................................... 21
Materials and Methods ............................................. 23
Results and Discussion ............................................ 28
Conclusions ..................................................... 53

APPENDIX A

SUBSTRATE DATA BASED ON THREE CORES

TAKEN FROM HYPORHEIC CHAMBERS ................................. 57

APPENDIX B

EXAMPLE OF CALCULATIONS FOR EACH METHOD

USING DATA FROM AUGUSTA CREEK ON 29 JULY 1997 .................. 59

LITERATURE CITED .................................................. 65

Table 1.

Table 2.

LIST OF TABLES

A comparison of values determined with the ﬂow-through hyporheic
chamber with values obtained for Bott et al. standard benthic chambers.
Values are means and standard errors for community respiration (CR24),
gross community productivity (GCP) , and P/R values for sites on four
Michigan Rivers ............................................ 12

Analysis of variance, blocking for the effects of river and date, for
community respiration (CR) and gross community productivity (GCP)
calculated using SYSTAT version 5.0. CR and GCP measured using
hyporheic chambers were signiﬁcantly different (p < 0.05) from values
measured with closed benthic chambers. Hyporheic chambers estimated
higher CR and GCP ......................................... 13

vi

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

LIST OF FIGURES

A. Design of the one m ﬂow-through, in-situ hyporheic chamber for
measuring community metabolism. B. The chambers were ﬁlled with
sediment using a customized shovel and cover designed to ﬁll the area
between sampling ports with two shovels full of sediments. The lid was
made of 0.6 cm clear acrylic and was sealed to the top of the chamber with
a foam gasket and secured with spring clamps. Sample compartments
were lined with 500 um mesh netting and were not ﬁlled with sediments.
Dissolved oxygen readings were taken from the sampling compartments
using YSI dissolved oxygen probes. Twelve-volt DC stirrers were placed
into the sampling compartments and directly beneath the probes (A). The
sediment transfer shovel (B) was used to ﬁll the hyporheic chambers. A
depression was made downstream of a ﬂat Shovel driven into the
streambed. The scoop was placed in the depression before removing the
ﬂat shovel and the scoop was driven upstream until it was full. The cover
was driven into the streambed to enclose the block of sediments (B) . . . . 6

After being ﬁlled with sediments (see ﬁgure 1), the hyporheic chamber
was buried in the streambed parallel to ﬂow. Chambers were left in place
with the covers off for a four-week incubation period before measurements
were made ................................................. 8

Gross community productivity (GCP) was measured using hyporheic and
benthic chambers simultaneously on six occasions. Results from
individual streams were compared using Mann-Whitney U tests. There
were signiﬁcant differences between chambers at the NE. of the AuSable
River and the Jordan River ................................... 14

Community respiration (CR24) was measured using hyporheic and closed
benthic chambers in a series of side-by-side experiments. Results were
compared using Mann-Whitney U tests. On every occasion, hyporheic
chambers estimated signiﬁcantly higher CR2, values ............... 15

P/R values were estimated using hyporheic and closed benthic chambers
simultaneously in a series of side-by-side experiments. Signiﬁcant
differences (p < 0.05) in measured values using each type of chamber were
observed on all occasions. Hyporheic chambers always predicted
heterotrophy while closed benthic chambers predicted autotrophy ..... 17

Net community productivity was estimated using the hyporheic chamber,
benthic chamber, and the upstream-downstream technique. Both the July
Augusta Creek site and the September Kellogg Forest Augusta Creek site
were monitored in 1997 and 1998. The upstream-downstream estimates

vii

Figure 7.

Figure 8.

Figure 9.

Figure 10.

Figure 11.

Figure 12.

Figure 13.

Figure 14.

Figure 15.

were made using two-Station calculations ........................ 29

Net community productivity was estimated using the hyporheic chamber,
benthic chamber, and the upstream-downstream technique. The upstream-
downstream estimates were made using single-station calculations. The
July upstream-downstream estimates become more variable from 1997 to

‘5

1998 when single-station calculations were used .................. 31

The replicates from 1997 and 1998 for each method were combined and
compared statistically using a Mann-Whitney U test (p < 0.05). Since the
upstream-downstream technique can not be replicated, n equals two for
both sites. Alike letters indicate signiﬁcant (p < 0.05) differences . . . . 32

Net daily metabolism was estimated using the hyporheic chamber, benthic
chamber, and the upstream-downstream technique. Both the July Augusta
Creek site and the September Kellogg Forest Augusta Creek site were
monitored in 1997 and 1998. The upstream-downstream estimates were
made using two-station calculations ............................ 35

Net daily metabolism was estimated using the hyporheic chamber, benthic
chamber, and the upstream-downstream technique. The upstream-
downstream estimates were made using single-station calculations . . . . 36

The replicates from 1997 and 1998 for each method were combined and
compared statistically using a Mann-Whitney U test (p < 0.05). Since the
upstream-downstream technique can not be replicated, n equals two for
both sites. Alike letters indicate signiﬁcant (p < 0.05) differences ..... 38

Community Respiration was estimated using the hyporheic chamber,
benthic chamber, and the upstream-downstream technique. Both the July
Augusta Creek site and the September Kellogg Forest Augusta Creek site
were monitored in 1997 and 1998. The upstream-downstream estimates
were made using two-station calculations ........................ 40

The replicates from 1997 and 1998 for each method were combined and
compared statistically using a Mann-Whitney U test (p < 0.05). Since the
upstream-downstream technique can not be replicated, n equals two for
both sites. Alike letters indicate signiﬁcant (p < 0.05) differences ..... 4]

Gross community productivity was estimated using the hyporheic
chamber, benthic chamber, and the upstream-downstream technique. Both
the July Augusta Creek site and the September Kellogg Forest Augusta
Creek site were monitored in 1997 and 1998. The upstream-downstream
estimates were made using two-station calculations ................ 45

The replicates from 1997 and 1998 for each method were combined and

viii

Figure 16.

Figure 17.

compared statistically using a Mann-Whitney U test (p < 0.05). Since the
upstream-downstream technique can not be replicated, n equals two for
both sites. Alike letters indicate signiﬁcant (p < 0.05) differences ..... 47

P/R values were estimated using the hyporheic chamber, benthic chamber,
and the upstream-downstream technique. Both the July Augusta Creek
site and the September Kellogg Forest Augusta Creek site were monitored
in 1997 and 1998. The upstream-downstream estimates were made using
two-station calculations ...................................... 49

The replicates from 1997 and 1998 for each method were combined and
compared statistically using a Mann-Whitney U test. Since the upstream-
downstream technique can not be replicated, n equals two for both

sites. Alike letters indicate signiﬁcant (p < 0.05) differences ......... 50

ix

Chapter 1
Including 20 cm of hyporheic respiration in

estimates of production to respiration

INTRODUCTION

Many studies concerning hyporheic nutrient dynamics suggest that an extensive
microbial community exists beneath the surface sediments (Grimm and Fisher 1984,
Rutherford and Hynes 1987, Ford and Naiman 1989, Triska et al. 1989a, Triska et al.
1989b, Triska et al. 1990, Hendricks and White 1991, Stanford and Ward 1993,
Hendricks and White 1995, Jones et al. 1995a). Until recently, this community has been
largely ignored when measuring ecosystem metabolism (P/R). Grimm and Fisher (1984)
suggested that the use of in-situ benthic chambers to estimate P/R values may
overestimate autotrophy by underestimating hyporheic respiration. While the closed
benthic chamber method, described by McIntire et al. (1964) and developed by Bott et al.
(1978), provides good estimates of surface sediment metabolism, it does not take into
account the potentially high respiration of the hyporheic zone. Pusch and Schwoerbel
(1994) speculated that respiration in many streams may actually be twice as high as
benthic chamber measurements suggest, if the hyporheic zone were included. Even
bottomless benthic chambers do not account for hyporheic respiration since surface and
interstitial waters do not readily mix once the chamber is put into place (Grimm and
Fisher 1984, Pusch and Schwoerbel 1994).

Currently the most popular methods for measuring community metabolism are

closed benthic chambers and diel oxygen curves or the upstream-downstream method.
Air-water diffusion estimates for the upstream-downstream method are Often imprecise
and add major uncertainty to P/R measurements obtained using this technique (Marzolf et
al. 1994). Closed system procedures are often favored because they eliminate the
uncertainty associated with estimates of diffusion. However, closed systems are not
without their Shortcomings as ambient conditions are altered to some extent. Altered
water velocity, temperature, nutrient concentrations, or dissolved oxygen (DO)
concentrations have the potential to change the metabolism of a closed sediment sample
(Homer and Welch 1981, Homer et al. 1990). An alternate procedure such as the
upstream-downstream method (Odum 1956) has the potential to account for hyporheic
respiration. Unfortunately, there are several problems associated with this method,
including the lack of true replication, the restriction of applicability to reaches with little
inﬂow, the uncertainty associated with oxygen ﬂux reaeration corrections, and the limits
of detecting small changes in D0 concentrations (Marzolf et al. 1994).

Grimm and Fisher (1984) estimated respiration of the hyporheic zone by sealing
both ends of a sediment core sample and monitoring DO consumption over time. Pusch
and Schwoerbel (1994) used a similar method, in which tubes ﬁlled with sediment were
buried parallel to ﬂow at various depths. The tubes were retrieved from the streambed
after an incubation period, but left submerged in the stream water to observe DO changes
under recirculating conditions. Naegeli et a1. (1995) used a similar system in the lab after
tubes were retrieved. Jones et al. (1995b) removed cores from the stream and ﬁlled them
with hyporheic water, sealed them and returned them to the hyporheic zone for a three to

four-hour incubation period. These four studies of hyporheic respiration indicated that

hyporheic respiration was similar to respiration measured at the surface on an areal basis.
Thus, hyporheic respiration should be included in estimates of community metabolism.
However, incubations of closed cores ex-situ may not produce representative values for
hyporheic sediments in-situ.

To address this problem, a ﬂow-through hyporheic chamber that allowed in-situ
determination of community metabolism was designed. The Objective of this study was
to test this newly designed hyporheic chamber that included surface and hyporheic
community metabolism in streams in side-by-side comparisons with the closed benthic
community metabolism chambers built to the design developed by Bott et al. in 1978.
The Bott et al. design has been used extensively since 1978 to measure P/R ratios for
streams.

The hypothesis tested was that inclusion of 20 cm of hyporheic sediments would
signiﬁcantly alter P/R values, because of increased community respiration by hyporheic
consumers. It was further hypothesized that inclusion of this hyporheic community
respiration would reduce P/R values for middle-order streams from autotrophic values

greater than one to heterotrophic values less than one.

MATERIALS AND METHODS

Study sites

Sites on the Jordan, AuSable, North Branch of the AuSable and the Red Cedar
Rivers in Michigan were selected for this study. Measurements were made at the Jordan
River, a designated wild and scenic river, on 5 September 1996. The Jordan was ﬁfth-
order, and heavily sand impacted at the study site located at T.3ON. R.6W. Sec. 10,
Antrim County, Michigan. The riparian zone at this site was forested and the stream
channel investigated was 18 m wide. Community metabolism using the side-by-side
comparison of methods was determined on the North Branch of the AuSable River on 16
May 1996 and 26 June 1997. Metabolic measurements were made on the main stem of
the AuSable River on 17 May 1996 and again on 17 July 1996. The AuSable River and
the North Branch AuSable River study sites were both located in Crawford County,
Michigan at T.26N. R2W Sec. 6 and T.27N. RIW. Sec 8 respectively. Both were
forested fourth-order stream reaches considered "quality trout ﬁshery stream reaches" by
the Michigan Department of Natural Resources. Stream width ranged from 15 to 20 m at
both sites and substrates consisted of mainly gravel. Metabolism measurements were
made on the Red Cedar River on 8 June 1997. The Red Cedar study site was a 3'd-order,
15 m wide reach of the river that drained predominantly urban and agricultural land. The
site was located at T.4N. R.1W Sec. 27 in Ingham County, several kilometers
downstream from a sewage treatment facility. Substrate consisted mainly of gravel with
an underlying, heavily consolidated clay layer. Unlike the other sites located on cold

water trout streams, the Red Cedar River was a warm water stream.

All measurements were made in rifﬂe sections except for measurements on the
Jordan River. Since rifﬂes did not exist on this stream reach, a comparable run of rapid
ﬂow was selected. Rifﬂe/run sections were chosen because they potentially support the

most primary productivity and are therefore most likely to be autotrophic.

Hyporheic chamber design

Flow-through, one m long, 20 cm wide, 20 cm deep, hyporheic chambers were
designed to include sediments that extended from the sediment surface to 20 cm deep.
Chambers were constructed from galvanized sheet metal and were covered with 0.6 cm
thick clear acrylic tops to allow light to penetrate to the sediments (Figure 1A). The
design allowed uninhibited parallel ﬂow through the entire length of the chamber.
Sampling ports at upstream and downstream ends were encased in 500 mm mesh to keep
sediments from ﬁlling around the stirrers and DO probes. Although the entire chamber,
upstream , downstream, and between the sampling ports was ﬁlled and buried, metabolic
measurements were made on sediments ﬁt in the 0.765 m long area between sampling
ports. Therefore, a sediment sample size of 0.765 m by 0.20 m by 0.20 m was used to
estimate commrmity metabolism.

With the top and bottom sealed, the ﬂow-through hyporheic chamber was not
subject to the uncertainty associated with estimates of air-water diffusion Of oxygen into
or out of the stream and/or with seepage of groundwater into the chamber between the
upstream and downstream ports. Thus, this chamber combines some of the best features
of the upstream-downstream technique of measuring stream metabolism with some of the

best features of the closed benthic chamber. The downside to this chamber is that it is

Clear acrylic top

 

 

ﬁr Sampling port lined with
_ - ’ /rneah
* x T . sediment

. ’ Flow
1 . .j ' /

 

 

 

 

 

 

 

 

 

1 D .0. mm! D 0 “m“ Galvanized thee! metal

 

 

 

 

 

 

 

 

I)

TIT _

l

l

. ’ l
/‘ L

/ / / / Sediments / / /

 

Figure l. A. Design of the one In ﬂow-through, in-situ hyporheic chamber for measuring
community metabolism. B. The chambers were ﬁlled with sediment using a customized
shovel and cover designed to ﬁll the area between sampling ports with two shovels full of
sediments. The lid was made of 0.6 cm clear acrylic and was sealed to the top of the
chamber with a foam gasket and secured with spring clamps. Sample compartments were
lined with 500 um mesh netting and were not ﬁlled with sediments. Dissolved oxygen
readings were taken {Tom the sampling compartments using YSI dissolved oxygen
probes. Twelve-volt DC stirrers were placed into the sampling compartments and
directly beneath the probes (A). The sediment transfer shovel (B) was used to ﬁll the
hyporheic chambers. A depression was made downstream of a ﬂat shovel driven into the
streambed. The scoop was placed in the depression before removing the ﬂat shovel and
the scoop was driven upstream until it was full. The cover was driven into the streambed
to enclose the block of sediments (B).

measuring metabolism on a very small spacial scale and the chamber itself can potentially
alter natural ﬂow paths since it is buried in the streambed. The chamber was made to
measure a conservative 20 cm of hyporheic zone for the purpose of this study, but within

reason, the size of the chamber is not limited by its design.

Filling Chambers

Four chambers were ﬁlled with 20 cm of sediment using a customized shovel and
cover to maintain the stratigraphic integrity of the sediments (Figure 1B). The outer
dimensions of the shovel matched the inner dimensions of the chamber, and each
hyporheic chamber held two shovels of sediment. Once the covered shovel was lowered
into the chamber, sediments were held in place as the shovel was removed from the
chamber. Hyporheic chambers were buried in the streambed parallel to surface ﬂow with
the covers removed, and allowed to incubate for four weeks (Figure 2). Due to the
unstable shifting sands of the Jordan River, the incubation period for both the hyporheic

and benthic chambers was eliminated.

Metabolism measurement

After incubation, 12-volt DC stirrers were placed into the sampling compartments
of three randomly chosen chambers. The chambers were covered by seating the lid
gaskets in the grooved edge of each chamber and attaching spring clamps. Yellows
Springs Instruments dissolved oxygen probes were securely sealed into the lid
corresponding to the upstream and downstream sampling compartments of each chamber.

Each replicate pair of probes was calibrated simultaneously before securing them in

Stream Flow

 

t D.O.Probes

I ,' L—«r'h‘; ”3 ' r—
\J W“

“E Siibs

/‘_é

92W?” .
urface ﬂow " -—'

 

Figure 2. After being ﬁlled with sediments (see ﬁgure 1), the hyporheic chamber was
buried in the streambed parallel to ﬂow. Chambers were left in place with the covers off
for a four-week incubation period before measurements were made.

sampling compartments.

The fourth chamber was used to determine hyporheic ﬂow velocity by introducing
a conductivity spike in the upstream sampling compartment with a concentrated sodium
chloride solution. Velocity measurements were conducted in a separate chamber in order
to avoid deleterious impacts to the biota. Velocity through all of the chambers was
assumed to be comparable since the relatively small sediment samples for all four
chambers were randomly taken from similar Sized sediments from the same rifﬂe.
Preliminary observations, where ﬂows through all four chambers were determined with
conductivity spikes, during the development of the chamber design agreed with this
assumption. A conductivity probe was used to detect the conductivity spike as it moved
through the downstream sampling compartment. Mean water velocity was calculated
from the time required for the conductivity spike to travel from the upstream to the
downstream compartment.

Dissolved oxygen concentrations in the upstream and downstream compartments
were monitored hourly for a 24-hour period. Stirrers were connected to a 12-volt DC
battery until dissolved oxygen meters stabilized. Immediately following each reading,
stirrers were disconnected. The interval of time determined by the sodium chloride tracer
was used to separate upstream and downstream readings to ensure a measurement of the
same water mass. The rate of change of oxygen between the upstream and downstream
compartments was calculated using the formula modiﬁed from McDiffett et a1. (1972):

(l) X = AC (F/A), where

X = rate of change (g 02 111‘2 hr")
C = dissolved oxygen (mg I")

F = discharge (m3 hr")
A = area of surface sediments (m2)

The upstream—downstream calculations were applied to the chamber since it is
essentially a small scale upstream-downstream technique without diffusion. Like the
upstream-downstream method, results from the hyporheic chamber were expressed per
meter square of surface sediment to compare results with the closed benthic chamber.
Discharge through the sediments was determined using water velocity, the cross-sectional
area of the chamber, and a conversion factor estimating the interstitial spaces of the
sediment. The conversion factor was calculated from an estimate of the sediment
porosity, the volumetric (V/V) ratio of water to sediment. Three sediment samples from
each site were dried at 105° C for 24 hours, and porosity was determined via volume
displacement (Fetter 1994). The area occupied by the interstitial spaces was determined
by multiplying the mean ratio of water to sediments by the chamber volume. Net
community productivity (N CP) was determined by calculating the mean rate of change
during the daylight period. Community respiration (CR24) was determined by calculating
the mean rate of change from sundown to sunrise and extrapolating to a 24-hour period.
Gross community productivity (GCP) was determined by multiplying CR per hour by the
number of daylight hours and adding the NCP. Community metabolism (P/R) was
determined by dividing GCP by CR24 (Bott et al. 1978).

Hyporheic chambers were used Simultaneously with standard chambers, as
described by Bott et al. (1978) and Bott in Hauer and Lamberti (1996). Results produced
by each type of chamber were compared using analysis of variance. Chambers were also
compared on an individual stream basis using Mann-Whitney U tests. The software

package SYSTAT 5.0 was used for the analysis.

10

RESULTS

Gross community productivity

GCP values (i SE) obtained from the ﬂow-through hyporheic chambers ranged
from 1.57 i 0.812 g 02 In2 day'I at the AuSable River during the spring to 4.39 :t 0.29 g
02 m“2 clay'l at the North Branch of the AuSable during midsummer (Table 1). Closed
benthic chamber estimates were lower and ranged from 0.78 :t 0.04 g 02 m'2 day" at the
AuSable River in the spring to 2.11 i 0.14 g 02 m'2 day'1 at the Jordan River during the
early fall. Measurements of GCP were signiﬁcantly different between chambers on three
of the six occasions with the hyporheic chamber always higher (Figure 3). The general
trend from site to site appeared to be similar between chambers with the exception of the
North Branch of the AuSable River. The hyporheic GCP values observed at the North
Branch of the AuSable River were the highest values measured over the course of the
investigation. This was not the case with the benthic chamber. When all sites were
combined for analysis of variance, blocking for river, there was a signiﬁcant difference (p
< 0.05) in GCP between the two chamber types (Table 2). The ﬂow-through hyporheic
chamber measured higher GCP on every occasion, although variance between replicates

was also higher.

Community respiration
Mean CR24 (:t SE) measured with the ﬂow-through hyporheic chamber ranged
from 32.75 1: 22.71 g 02 m'2 day'1 at the AuSable River during the spring to 7.13 :h 0.13 g

02 m’2 day'l at the Red Cedar River in early summer (Table 1).

ll

Table 1. A comparison of values determined with the ﬂow-through hyporheic chamber
with values obtained for Bott et al. Standard benthic chambers. Values are means and

standard errors for community respiration (CR24), gross community productivity (GCP) ,
and P/R values for sites on four Michigan Rivers.

20 cm deep hyporheic chambers with flow through the sediments

CR,, standard GCP
(g 02 m'2 day") error

standard P/R standard
(g 02 m'2 day") error (GCP/CR24) error

 

AuSable R. 32.75 3:22.71 1.57 $0.812 0.06 $0.01
(5/17/96) n=3

AuSable R. 11.32 $2.47 2.44 10.61 0.21 $0.07
(7/17/96) n=2
N.B.AuSable 23.18 $7.6 4.39 $0.29 0.22 $0.05
(7/ 16/96) n=3

Jordan R. 9.60 $2.39 3.58 :1:0.60 0.42 $0.13
(9/5/96) n=3
N.B.AuSable 19.89 111.23 3.82 11.12 0.26 $0.07
(6/26/97) n=3
Red Cedar R. 7.13 $0.70 2.97 i080 0.41 :00?

(6/8/97) n=2

Bott et al. (1978) recommended benthic chambers with flow over the sediments

 

 

CR2, standard GCP standard P/R standard
(g O2 rn‘2 day") error (g 02 at2 day") error (GCP/CR2.) error
AuSable R. 0.65 $0.04 0.78 £0.04 1.2 £0.05
(5/ 17/96)
n=5
AuSable R. 1.45 $0.19 1.43 $0.09 1.02 $0.1 I
(7/ 17/96) n=4
N.B.AuSabe 1.00 i004 1.34 $0.12 1.34 $0.14
(7/16/96) n=4
Jordan R. 0.57 $0.05 2.11 $0.14 3.78 $0.52
(9/5/96) n=3
N.B.AuSabe 0.94 $0.22 1.1 1 $0.33 1.16 10.06
(6/26/97) n=3
Red Cedar 0.63 10.23 0.96 i039 1.56 10.21

(6/8/97) n=3

12

Table 2. Analysis of variance, blocking for the effects of river and date, for community
respiration (CR) and gross community productivity (GCP) calculated using SYSTAT
version 5.0. CR and GCP measured using hyporheic chambers were signiﬁcantly
different (p < 0.05) from values measured with closed benthic chambers. Hyporheic
chambers estimated higher CR and GCP.

Dependent Variable: Community Respiration n=38 Multiple R: 0.631 Squared
Multiple R: 0.399

 

 

Source Sum-of-Squares DF Mean-Square F -Ratio P
(Type 111)
Block 0.894 5 0.179 0.635 0.675
Chamber 4.948 1 4.948 17.567 0.000
Error 8.731 31 0.282

 

Dependent Variable: Gross Community Productivity n=38 Multiple R: 0.802
Squared Multiple R: 0.643

 

 

Source Sum-of-Squares DF Mean-Square F-Ratio P
(Type 111)
Block 11.520 5 2.304 2.851 0.031
Chamber 29.813 1 29.813 36.888 0.000
Error 25.054 31 0.808

 

Values were much lower when measured with the closed benthic chamber,
ranging from 1.45 i 0.19 g 02 m'2 day'1 at the AuSable River during midsummer to 0.57
a: 0.05 g 02 m’2 day’1 at the Jordan River in early fall. The hyporheic chamber measured
CR2, signiﬁcantly higher on every occasion and like GCP measurements, variance was
also much higher (Figure 4). There did not seem to be a similar general trend between
chambers from site to site although this would be difﬁcult to detect because it may be

masked by the variance of the hyporheic chamber. When all sites were combined for

13

Gross Community Productivity

 

 

 

 

 

 

 

 

 

 

 

 

5 .-
L * ‘1.
4
I. * g * P
a
3 ————— -
‘7‘ ..
E qr
§ .
8' 2 1 r
9’ in
Q §
1 l I
l o I T
ad ad") 90 go“ 66 669‘
, . O Hyporheic Chamber
p<0.05 (Mann-Whitney U) o Benthic Chamber

Figure 3. Gross community productivity (GCP) was measured using hyporheic and
benthic chambers simultaneously on six occasions. Results from individual streams were
compared using Mann-Whitney U tests. There were signiﬁcant differences between
chambers at the NE. of the AuSable River and the Jordan River.

14

Community Respiration

 

 

 

 

 

 

 

 

 

 

 

 

T
50
40
F> 0 .
8 30
cis
C5“ 0 ' .
20
U)
10 i t I " .
I a
0 n O O 3 Q
\0 \0 \0 30 \O 9‘
we so $9 W9 5° $9 1*" 9.060

a Hyporheic Chamber

- p<0.05 (Mann-Whitney U. one'tai'ed> o Benthic Chamber

Figure 4. Community respiration (CR24) was measured using hyporheic and closed
benthic chambers in a series of side-by-side experiments. Results were compared using
Mann-Whitney U tests. On every occasion, hyporheic chambers estimated signiﬁcantly
higher CR2, values.

15

analysis of variance, blocking for river, there was a signiﬁcant difference (p < 0.05) in

measured CR24 between the two chambers (Table 2).

P/R values

P/R values always suggested heterotrophy when measured with the ﬂow-through
hyporheic chamber. Values ranged from 0.42 at the Jordan River in early fall to 0.06 at
the AuSable River during spring (Table 2). Conversely, benthic chambers predicted
autotrophy on every occasion with values ranging from 3.78 at the Jordan River in early
fall to 1.02 at the AuSable River during midsummer. P/R values were signiﬁcantly
different (p < 0.05) on every occasion (Figure 5). The two chambers seemed to produce
similar trends with respect to P/R values from site to site with the exception of the

AuSable River during the summer.

16

PIR Values

 

 

 

 

 

 

 

 

 

 

T
4.0
Q?
3.5
.11 J
z r
s 7 I”
n. i
1.5 * 1? 1
Q *Q
1 o * R
- I—
0.5 : 3.-“
o I 1 i
0.0 ~—!
\6 \0 \e a“ “\0 60‘
5’6“ 59“ 50° 0(9 :33 06
P9 P9 93 1*" 3 $9. 0 9B9
* p<0.05 (Mann-Whitney u. one-tailed) ; SgﬁgﬂeghEQaT‘fe'

Figure 5. P/R values were estimated using hyporheic and closed benthic chambers
simultaneously in a series of side-by-side experiments. Signiﬁcant differences (p < 0.05)
in measured values using each type of chamber were observed on all occasions.
Hyporheic chambers always predicted heterotrophy while closed benthic chambers
predicted autotrophy.

DISCUSSION

Gross community productivity

Estimates of GCP produced by each type of chamber were signiﬁcantly (p < 0.05)
different on three of the six occasions. When the Sites were combined for analysis of
variance, there was a signiﬁcant (p < 0.05) difference between chambers. Hyporheic
chamber estimates of GCP were higher on every occasion. However, differences between
chamber estimates of GCP did not increase hyporheic chamber P/R estimates with respect
to Benthic chamber estimates as might be expected. In fact, higher GCP estimates
measured using the hyporheic chamber still resulted in signiﬁcantly lower P/R estimates.
Therefore, the hyporheic chamber estimates of P/R were conservative with respect to
autotrophic benthic chamber estimates since the hyporheic chamber measured higher

GCP.

Community respiration

CR was always at least an order of magnitude higher when measured with the
ﬂow-through hyporheic chamber. These measurements only included 20 cm of the
hyporheic zone. Many studies suggest that the hyporheic zone often extends to much
greater depths than 20 cm in many streams (Grimm and Fisher 1984, Rutherford and
Hynes 1987, Ford and Naiman 1989, Triska et al. 1989a, Triska et al. 1989b, Triska et al.
1990, Hendricks and White 1991, Stanford and Ward 1993, Hendricks and White 1995,
Jones et a1. 1995) which may make these estimates of CR24 extremely conservative. The

estimates of CR2, at the Red Cedar likely included the entire hyporheic zone. A uniform

l8

layer of heavily consolidated clay was found below ﬁve cm in all sediment samples used
in hyporheic chambers at the Red Cedar. These data support the fact that the hyporheic
zone did not contribute as much to CR24 at this site since CR24 measured using hyporheic
chambers was lowest at this site even though it undoubtedly had the most associated
anthropogenic disturbance and probably the highest nutrient concentrations. The
pronounced differences in CR2, between the two chamber types may be an indication of a
very active but not extensive hyporheic zone. Evidence provided by this study and
Mulholland et al. (1997) indicates that the extent of the hyporheic zone seems to be an

important determinant of CR2, within stream reaches.

P/R values

On every occasion hyporheic chambers predicted heterotrophy (P/R <1) while
benthic chambers predicted autotrophy (P/R >1). GCP and CR2., measured with the
hyporheic chambers showed a great deal of variance between replicates within a single
rifﬂe. When the two parameters were put together as P/R values the variance was greatly
decreased, this remained true, even when an arcsine squareroot transformation was used.
It was noted that when a single replicate showed high GCP, CR24 was also high. The
converse was also true when GCP was low for a given chamber. Jones et a1. (1995) also
concluded that hyporheic respiration was tightly linked to surface production. This may
be due to leaching of streamwater DOC, as a result of surface productivity, into the
sediments locally stimulating hyporheic respiration.

These data suggest that even relatively unirnpacted medium size alluvial rivers in

Michigan may be heterotrophic. This observation leads to questions concerning the

19

reliability of current predictions regarding P/R ratio changes along a river continuum, and

emphasizes the need to include the hyporheic zone in studies of stream metabolism.

20

CHAPTER 2
Metabolism in a 3'“-order stream:

a comparison of Methods

INTRODUCTION

Closed benthic chambers as described by Bott et al. (1978) and diurnal oxygen
curves or the upstream-downstream method (Odum 1958) are both widely accepted
methods of measuring stream metabolism. The two methods have often been used
interchangeably. In fact, the study reach of interest or availability of specialized
equipment often dictates which method is used (Bott 1996). Benthic chambers are often
favored where lateral inﬂow is high, oxygen diffusion across the air-water interface is
expected to be high, or an indication of variance is required. The upstream-downstream
method may be favored where a range of habitat types are to be incorporated in the
measurement, or a nutrient limitation is expected to occur if closed benthic chambers are
used. Unfortunately, the two methods can give conﬂicting results when used in
conjunction (Bott et al. 1978; Grimm and Fisher 1984; Pusch and Schwoerbel 1994). In
light of recent ﬁndings of signiﬁcant respiration in the hyporheic zone, it is not surprising
that results would conﬂict. After all, the two methods measure different portions of the
stream. Bott et al. (1997) suggest selecting a method most consistent with the
experimental objectives ranging from micro- to macro-habitats.

The upstream-downstream method should theoretically include hyporheic

respiration so long as there is return ﬂow from the hyporheic zone within the study reach.

21

However, the upstream-downstream method has been criticized because it relies heavily
on estimates of reaeration and much uncertainty accompanies these estimates. Recently
improvements have been made to the upstream-downstream method. Marzolf et al.
(1994) used both a conservative and volatile tracer, thus increasing conﬁdence associated
with reaeration calculations. In an attempt to produce an indication of variance
associated with this method, McCutchan et al. (1998) suggested the use of a Monte Carlo
approach to estimate the overall uncertainty by incorporating error associated with each
individual measurement made to calculate metabolic parameters.

I developed a chamber technique that accounts for streambed metabolism
including hyporheic respiration to a depth speciﬁed by the dimension of the chamber
(chapter 1). The purpose of this experiment was to compare the results obtained
simultaneously using the closed benthic chambers recommended by Bott et al. (1978)
with results from the hyporheic chambers that I developed, and with results obtained
using the upstream-downstream technique with the improvements recommended by

Marzolf et al. (1994).

22

MATERIALS AND METHODS

Study Site

The study sites were located on the third-order Augusta Creek in Kalamazoo
County, Michigan. The site studied in July 1997 and July 1998 was located at T. IS. R.
9W. Sec 10. The stream is considered to be relatively pristine for the ecoregion. The 35
m reach of stream was approximately 10 m wide with a forested riparian zone providing a
fairly dense canopy. The mean water depth of the reach was 0.10 to 0.15 m and ﬂowed
over a gravel/pebble streambed.

The Kellogg Forest site studied in September 1997 and September 1998 was
located at T. 1S. R. 9W. Sec 22. This 30 m reach was approximately 10 m wide and also
had a forested riparian zone, however, the canopy was not as dense as the site studied in
July. The reach could be described as a very sandy low gradient run. The mean depth

was approximately 0.41 m.

Methodology

The hyporheic and benthic chamber methods were used as described in chapter I.
I used the upstream-downstream method slightly modiﬁed from Marzolf et al. (1994).
Dissolved oxygen (D.O.) probes were placed in the thalweg and at the upstream and
downstream ends of the study reach. Downstream D.O. concentration were measured one
reach travel time following each upstream measurement. Measurements were made
hourly for 24 hours. The difference between the upstream and downstream

concentrations are thought to be caused by dilution, chemical oxygen demand,

23

metabolism and reaeration ﬂux of oxygen. Rhodamine-WT, a conservative tracer,
Showed that dilution was insigniﬁcant within the reach. Lateral inﬂow was less than 1%
in either study reach. Chemical oxygen demand was considered to be insigniﬁcant
leaving only metabolism and reaeration to account for changes in DC. Therefore, hourly
readings were corrected for reaeration using calculations developed with a continuous
injection of a conservative tracer and propane, a volatile tracer of air-water gas exchange.
Rhodamine-WT was used to determine in-stream ﬂow rate, lateral inﬂow, and the
reach water travel time (r). Reach water travel time determined the sampling interval
between the upstream and downstream stations, and the length of time required for
propane concentrations to reach steady-state (four times 1:). A solution of 1.14 g/L
Rhodamine-WT was dripped into the stream from an 18-L Mariotte bottle at four ml/sec
150 m upstream of the sample reach. Water samples were taken simultaneously at the
upstream and downstream stations every 15 seconds beginning at the time of injection.
Sampling continued until steady- state was achieved. Grab-samples were collected in six-
dram glass sample vials at six tenths the depth and in the thalweg. Samples were left
standing until sediments settled and a constant temperature was achieved before they
were measured using a Turner Designs ﬂuorometer. Percent lateral inﬂow was
determined as the percentage decrease in dye concentrations between upstream and
downstream sites at steady-state . Reach water travel time was determined by the
difference in half-height concentration at steady-state. Discharge measurements were
made using a Marsh-McBirney ﬂow meter. The Rhodamine-WT procedure was carried
out one to two days before the upstream-downstream technique was administered during

base-ﬂow conditions between storm events, when Stream depths were relatively constant

24

from day to day. Thus, reach travel time was assumed to have changed very little from
the time the Rhodamine-WT procedure was carried out to the time of the community
metabolism study.

Propane was bubbled into the stream from a 20-lb tank containing HD-S (the
conunon commercial propane containing 96% propane and 4% ethane) through two-m
long perforated hoses. Most of the ethane was bled off before starting. Propane was
injected for four times 1: to reach steady-state concentration before water samples were
collected. Water samples from the upstream and downstream sites were collected one I
interval apart. This ensured that the same water mass was sampled, guarding against
ﬂuctuations in injection rates. Grab samples were collected using horizontal bottle
samplers. Water was transferred to 125-ml Boston round screw-cap bottles, overﬂowing
the bottle three times its volume. Bottles were then sealed using Open-hole caps equipped
with a TFE/silicone liner. Samples were immediately taken to the lab and propane
concentrations were determined using gas chromatography. An eight-ml helium
headspace was added to each sample before they were shaken to allow dissolved propane
to equilibrate with the helium headspace. A portion of the headspace was injected into a
gas chromatograph equipped with a ﬂame ionization detector, using helium as a carrier.
The integrated areas of the propane peaks were used directly in the reaeration coefﬁcient
(RC) calculations, instead of actual concentrations. The conservative and volatile tracers
were not injected into the stream simultaneously. The following equation was used to

calculate the propane exchange coefﬁcient,

(1) can): km... (mint) = r' * 1n {(6. »= C2)/(Gz * Ca}

25

where T is the water travel time through the rifﬂe, G1 is the steady-state concentration of
propane at the upstream station, G2 is the steady-state concentration of propane at the
downstream station, Cl is the steady-State concentration of the conservative tracer at the
upstream station, and C2 is the steady-state concentration of the conservative tracer at the

downstream station. The oxygen exchange coefficient (k ) was calculated as

oxygen
(2) koxygen = 1.39 * kpmpm. Rathbun et al. (1978) demonstrated that the relationship
between kpropane and koxygen is independent of mixing conditions, temperature, and water
quality.

The instantaneous oxygen ﬂux into or out of the stream is calculated using,

(3) Reaeration ﬂux = DOdeﬁcit * -k * 1: (Young and Huryn 1998; Marzolf et al. 1998)

om...
where koxygen = RC (min"), DOdeﬁcit = DOSmmwd - DOmmmd (mg of 02 per L), and 1: =
stream water travel time (min). The DOdeﬁcit is the difference between the calculated
oxygen saturation value and the mean of the upstream and downstream measurements.
Saturation value was calculated using the water temperature and the atmospheric
pressure.

The following day, in-stream DO concentrations were measured using YSI
oxygen meters and polarographic sensors. The meters were air calibrated before being
placed side by side in the stream where ﬁne adjustments were made to ensure that each
meter read identical concentrations. Probes were fastened to stakes placed in the thalweg
at the upstream and downstream stations. Time, temperature, and dissolved oxygen
concentrations were recorded hourly for 24 hours. DOSmumml was calculated using the

average stream temperature, barometric pressure, and elevation. The DOM“, was

calculated as the difference between DOS,“mad and the averaged upstream and

26

downstream concentration. The following metabolic parameters were calculated:

Dark metabolism = ADOdark - reaeration; Net community productivity (N CP) = ADOhgh,
- reaeration; Gross community primary productivity (GCP) = NCP + respirationhgh, ;
Community respiration (CR24) = Average night Resp. sealed for 24 hours; Net daily
metabolism (NDM) = GCP - CR24; and P/R = GCP/CR3. The change in DC. (ADO)

was calculated as the difference between the upstream and downstream values.

27

RESULTS AND DISCUSSION

Net Community Productivity

All three techniques directly measured net community productivity (N CP) during
daylight hours. Using oxygen as a surrogate for photosynthesis, NCP can be deﬁned as
the amount of oxygen either produced or consumed during the period from sunrise to
sunset. This parameter is free from error associated with extrapolating either gross
production or 24-hour respiration estimates. Every data point recorded during the
daylight period using the closed benthic chamber was a direct measure of NCP for a one-
hour period. Data points recorded using the upstream-downstream and hyporheic
chamber techniques were instantaneous NCP readings and were used to represent an
average for a particular hour.

Net community productivity estimates made in 1998 were very similar to
estimates made in 1997 for each method (Figure 6). The greatest difference between
1997 and 1998 was observed using the upstream-downstream technique at Kellogg Forest
in September. This difference may be due to increased error associated with this method
when it is reduced to a single station. The 1997 September upstream-downstream method
had to be reduced to a single station technique since there was not a signiﬁcant difference
between upstream and downstream dissolved oxygen concentrations. Differences were
detected using a paired t-test at that site in 1998 and therefore, the two-station technique
was employed as recommended by Bott (1996). Two-station upstream-downstream
calculations were possible during both July 1997 and 1998 measurements.

I re-calculated all of the NCP values using the single-station equations and found

28

9 02 m'2 clay'1

Net Community Productivity

(two-station calculations)

 

 

 

 

 

5
o
O _. .................................. . .9. ............. C5 ......................... C. O.
-5- ........................................................................................
o
-10- ........................................................................................
-15-- ...................................................................................
o
-20 o
’\ ’\ ’\
«9.9% 3‘: 9‘" «ﬁrst we 099 e" golf-9%
o ~e'x 9 'e'\
0 Q Q

o ‘l 997
o 1998

Figure 6. Net community productivity was estimated using the hyporheic chamber,
benthic chamber, and the upstream-downstream technique. Both the July Augusta Creek
site and the September Kellogg Forest Augusta Creek site were monitored in 1997 and
1998. The upstream-downstream estimates were made using two-station calculations.

29

that the standard error around these means increased (Figure 7). Calculating both 1997
and 1998 September data sets using the single-station equations instead of one calculated
as a two-station and the other as a single-station reduced the difference between these two
dates. This reduction is likely an artifact of using identical equations. The July
comparison suggests that standard error around the September mean may have been
reduced if two-station calculations were possible for both data sets.

I combined the replicates from both years for each method to statistically explore
differences in methodology. I justiﬁed this because the 1997 and 1998 measurements
within methods were very similar at both sites (Figure 8). Measurements from one year
to the next were so similar that within-method variance was actually reduced when the
years were combined. Caution should be used when analyzing results concerning the
upstream-downstream method because the upstream-downstream method has an n value
of only two. Unfortunately statistical analysis is not otherwise possible since the method
cannot be replicated and the Monte Carlo approach of McCutchan et al. (1998) was not
used.

The NCP values obtained using the upstream-downstream and hyporheic chamber
techniques were not signiﬁcantly different during the July (1997 and 1998 combined)
experiment. Both methods produced signiﬁcantly lower NCP values than the closed
benthic chambers. The benthic chambers suggested that NCP was slightly greater than
zero while the other two methods provided negative values. This is not surprising
because the latter two methods include hyporheic respiration.

The lack of a signiﬁcant difference between hyporheic chambers and the upstream-

downstream techniques suggests that the mean depth of the hyporheic zone at the July

30

Net Community Productivity

(Single-station calculations)

 

 

 

 

 

5
C
0 _, ................................... .. ................ (5 ......................... .. O
i O
‘—
'> -5 .1 ...................... Q ................................................................ I
m
'0
N O

I
E -1 0_ ............................................. . ...........................................
8' .
U,

-15-- ...................................................................................

-20

A ‘b A A
«$93, «@«F‘b «W9 @399?) dqd‘gb 019$)
*QO‘QO 0°00 @OQ'G’ *QO‘\ 0°00 ©o©o
’ ’ 0 0 ’ ’ e e
5?‘ \" QQQQ 9e, 90 ‘1‘ r?‘ 909 0g 90
o 1997
o ‘l 998

Figure 7. Net community productivity was estimated using the hyporheic chamber,
benthic chamber, and the upstream-downstream technique. The upstream-downstream
estimates were made using single-station calculations. The July upstream-downstream
estimates become more variable from 1997 to 1998 when single-station calculations were
used.

31

9 02 m’2 day"1

Combined 97I98 Estimates Of
Net Community Productivity

 

 

 

 

 

 

5
a b a
0 _ .................................. 9. ................. .a. ....................... 0
é o
-5 - .......................................................................................
-10- ..........................................................................................
a
-15-...f ..................... b. ............... . .............................................
-20
«9% «9e «9% (99% (59% 99%
«a «a 4b «a .(‘b «a
e o e e o 9
°«\ «\ 6“ q\ Oct ‘00}
Q . ‘ Q 9 ‘
.38 0Q 90° .5) 0Q 00°

Figure 8. The replicates from 1997 and 1998 for each method were combined and
compared statistically using a Mann-Whitney U test (p < 0.05). Since the upstream-
downstream technique can not be replicated, n equals two for both Sites. Alike letters
indicate signiﬁcant (p < 0.05) differences.

32

(1997 and 1998 combined) Augusta Creek site within the 36 m rifﬂe section was
approximately the same or slightly greater than the depth of the hyporheic chamber, or
that the majority of the respiration took place in the top 20 cm of the hyporheic zone and
a small percentage took place at depths greater than 20 cm. The large difference between
the benthic chamber and the other two methods suggests that inclusion of hyporheic
respiration signiﬁcantly reduces NCP at this site. The benthic chamber measured
approximately equal amounts of primary production and respiration at the streambed
during the daylight hours. The inclusion of hyporheic respiration revealed an oxygen
consumption rate during the daylight hours of approximately 15 to 20 grams per square
meter per day.

Similar results were expected at the Kellogg Forest site in September (1997 and
1998 combined), but this was not the case. Similar to the combined July experiments,
there was no signiﬁcant difference between the upstream-downstream and hyporheic
chamber techniques, and the hyporheic chamber produced signiﬁcantly lower NCP values
than the benthic chamber, but this time the upstream-downstream and benthic chamber
techniques were not signiﬁcantly different. All three of the methods produced NCP
values very near zero at this site.

Between-method NCP values were very similar, and therefore, these data suggest
that the mean depth of the hyporheic zone at Kellogg Forest was very small, likely only a
few centimeters. Previous studies showed that water does not circulate through the
sediments inside benthic chambers, so measurements of oxygen ﬂux in the chambers
reﬂect little or no hyporheic respiration. Particle size data supports the idea that the

hyporheic zone was very small at this site. Fine surﬁcial sediments in this 30 m run likely

33

created anaerobic conditions just below the streambed.

Net Daily Metabolism

Net daily metabolism (NDM) is deﬁned as gross community productivity (GCP)
minus 24-hour community respiration (CR24). Estimates Of both GCP and CR2, involves
extrapolating night-time respiration through daylight hours and therefore both include
error associated with this extrapolation. Algal respiration in the light is not the same as it
is in the dark, and photorespiration, or the enzymatic consumption of oxygen during the
synthesis of glycolate, cannot be accounted for (Bott 1996). When CR24 is subtracted
from GCP, the extrapolated respiration is removed. NDM is therefore free from errors
associated with extrapolating respiration during the day and only expresses direct
measures of NCP and night-time respiration.

NDM results followed similar patterns as NCP. Like NCP, within method
measurements were very similar between 1997 and 1998. This indicates that not only
were within method measurements made during the daytime similar, as shown by NCP
values, but also nighttime measurements were similar between 1997 and 1998 (Figure 9).
Again the greatest difference between years was observed using the upstream-
downstream technique at Kellogg Forest and can likely be attributed to the differences in
single-station and two-station calculations. The difference between years at Kellogg
Forest using the upstream-downstream method is reduced when both measurements are
calculated as single-station methods (Figure 10), but the two-station method is considered
to be more reliable and should be used when possible. When the replicates from both

years at the July August Creek site were combined, the hyporheic chamber and upstream—

34

Net Daily Metabolism

(two-station calculations)

 

 

 

 

 

O
O _ .................................. ' .Q . .. ......... (.3. ......................... . O
,7 -10- .................................. Mi ....................................
>
(U
'5 o
‘1‘
E -20- ........................................................................................
m %
-30- .................. ‘ ...................................................................
O
«93$ «VS/19% «VAR 3%” 3%" $565
9° 0%“ $6 o°<>° 6’s"
Qt) QsA 0Q Gd 600 00 Q3 .38 Q 9’ 000000
0 1997
o 1998

Figure 9. Net daily metabolism was estimated using the hyporheic chamber, benthic
chamber, and the upstream-downstream technique. Both the July Augusta Creek site and
the September Kellogg Forest Augusta Creek site were monitored in 1997 and 1998. The
upstream-downstream estimates were made using two-station calculations.

35

Net Daily Metabolism

(Single-station calculations)

 

 

 

 

 

 

O _. .................................... Q... ...................................... . O
. O
‘T
>.
m -10- ........................................................................................
'o o
‘7‘
E o
O" O
-20 _ ..." .....................................................................................
o: I L o
-30
«@193, «$92) «919% (8638033 ($35!: @3019?)
0 Q Q '0 '0 0 0 '0 '0
Q 9 ,°\ \ 9 9 \ \
3.495.480 09 0.; es 005‘ 6593590 09$ §§
e e e e
o 1997
o 1 998

Figure 10. Net daily metabolism was estimated using the hyporheic chamber, benthic
chamber, and the upstream-downstream technique. The upstream-downstream estimates
were made using single-station calculations.

36

downstream methods produced results that were not Signiﬁcantly different from each
other. Both were signiﬁcantly less than benthic chamber estimates (Figure l 1).

Neither the hyporheic nor the benthic chambers produced signiﬁcantly diffefent
NDM values from the upstream-downstream method at the September Kellogg Forest
site, but they were signiﬁcantly different from each other. The hyporheic chamber
showed signiﬁcant but only slightly lower NDM than benthic chambers at this site. The
relationship between measures of NDM among techniques was nearly identical to the
relationship between measures of NCP, except the difference, although not signiﬁcant,
between the hyporheic and the upstream-downstream methods appears slightly larger.
Since NDM can be reduced algebraically to NCP minus Crm-gm, it appears as if the
upstream-downstream method measured slightly more respiration than the hyporheic
chamber. Based on these un-extrapolated parameters, we can assume that the differences
between methods were only due to differences in measured respiration, or that differences
in respiration far outweighed any differences in measured productivity. These
conclusions are based on the fact that primary productivity is only measured during the
day while respiration is measured both day and night. Very similar patterns were
observed between NCP measured only during daylight and NDM measured both day and
night. Also, the most pronounced differences in NCP and NDM are between the benthic
chamber and the other two techniques at the July Augusta Creek site. This difference is

almost certainly due to difference in measured respiration. By design these methods

measure different amounts of CR.

37

g 02 m'2 day'1

Combined 97I98 Estimates of
Net Daily Metabolism

 

 

 

 

 

 

10
a b . a
o - .................................. . ........................................... . .
la L
-10- ............................................. 1 ............................................
..l.
-20- ..........................................................................................
E a
-30- .................. i . . . b ............................................................
«9° «9° «9° a" 99° 83°
«9 «b «13‘ «9 «9 «17
«\g «\q «\q 69 99 99
o“ *9 9° ,0“ '3‘
$0 69 0° 8“ 6Q e°°

Figure 11. The replicates from 1997 and 1998 for each method were combined and
compared statistically using a Mann-Whitney U test (p < 0.05). Since the upstream-
downstream technique can not be replicated, n equals two for both sites. Alike letters
indicate signiﬁcant (p < 0.05) differences.

38

Community Respiration

Community Respiration (CR24) is deﬁned as the amount of respiration measured
at night plus nighttime respiration extrapolated through the daylight hours. CR2, does not
account for differences in algal respiration in the light versus the dark.

Measures of CR24 made in 1998 using each method compared well to the previous
year (Figure 12). Again I combined replicates from each year for comparison. At the
July Augusta Creek site, the relationship of CR24 results between methods was the inverse
of the trend observed with NCP and NDM. This supports the idea that respiration was
the major factor controlling differences between methods. The upstream-downstream
method measured the highest CR24 at 38.0 mg 02 m'2 day" , the hyporheic chamber
followed with 24.6 mg 02 m'2 day" and ﬁnally the benthic chamber CR2, was 1.3 mg 02
In2 clay'I (Figure 13). All three methods produce results signiﬁcantly different from each
other.

The three methods used simultaneously can estimate the mean depth of the
hyporheic zone within the 35 m reach. CR2, associated with surface sediments was 1.3
mg 02 rn‘2 day" therefore, hyporheic respiration down to 20 cm Of depth was 23.7 mg 02
m’2 day", or 24.6 mg 02 111'2 day’1 minus 1.3 mg 02 m’2 day". I recommend subtracting
surface respiration because respiration associated with combined heterotrophs and
autotrophs is not likely comparable to heterotrophic respiration of the hyporheic zone.
While Fuss and Smock (1996) found that 70% of whole stream respiration occurred in the
hyporheic zone, the metabolic rate per surface area of substrate was lower than other
compartments. Other studies have found similar respiration rates at the streambed and in

the hyporheic zone on an aerial basis (Grimm and Fisher 1984; Pusch and Schwoerbel

39

Community Respiration

(two-station calculations)

 

 

 

 

 

50
4o - ...................... .0. ................................................................
.
30 - .........................................................................................
,-
u> £
% f
N 20 _ ........................................................................................
IE .
81 1o _ ............................................. . ..........................................
or
o 4 ................................... ...-9.... ......... F? .............. 0 ......... 0.0
«@456 «Rage, 39:16, 9133f 4359?, 96:59.6
0 8‘ '\ '\ 8‘ 0 '\ '\
«$80 683599 Qo‘iéf‘ Sﬁg‘fo o 099 Qa‘ioéo
o 1997
o 1998

Figure 12. Community Respiration was estimated using the hyporheic chamber, benthic
chamber, and the upstream-downstream technique. Both the July Augusta Creek site and
the September Kellogg Forest Augusta Creek site were monitored in 1997 and 1998. The
upstream-downstream estimates were made using two-station calculations.

40

9 02 m‘2 den/‘1

Combined 97I98 Estimates of

Community Respiration

 

 

 

 

 

 

50
a
40 _ .................. f ...................................................................
3o - ..-...a. ...............................................................................
2., i ...................................................................................
l a _
1o - ......................................................................................
a l " a
0 - .................................. .o. ............................ _, ............. . . . ..
«9% «9% «9% 69% 69% a‘f’
«if «9‘ «13' «‘6' «9‘ «‘9'
«9 «1° «\9 99 919’ 0)”
0“ x“ s s
it“? 69’ e°° 899° 099 e°°

Figure 13. The replicates from 1997 and 1998 for each method were combined and
compared statistically using a Mann-Whitney U test (p < 0.05). Since the upstream-
downstream technique can not be replicated, n equals two for both sites. Alike letters
indicate signiﬁcant (p < 0.05) differences.

41

1994; Naegeli et al. 1995; Jones et al. 1995). I assumed that hyporheic respiration is
relatively constant with depth and that the upstream-downstream method does account for
all of the hyporheic respiration. The mean depth of the hyporheic zone was
approximately 32 cm or 20 cm multiplied by 38.0 mg 02 m'2 day'l divided by 23.7 mg 02
m'2 day". The similarity of community respiration between years suggests that the depth
of the hyporheic zone remained relatively constant at the July Augusta Creek site from
1997 to 1998.

The between-method CR2, values measured at Kellogg Forest in September also
showed a pattern inverse of NCP and NDM when replicates were combined. There was
no signiﬁcant difference between the upstream-downstream method and the other two
methods. The hyporheic and benthic chambers did, however, produce small but
signiﬁcantly different results.

Benthic chambers produced nearly identical results from 1997 to 1998. CR24
measurements made with the upstream-downstream method and hyporheic chamber in
1998 were quite lower than they were in 1997. It appears as if the depth of the hyporheic
zone decreased from 1997 to 1998, and in fact, these data suggest that the hyporheic zone
was nearly non-existent in 1998. Possible explanation for the decrease would include
decreased stream water velocity through the reach increasing sedimentation, or an
increase in ground water upwelling. The latter is very unlikely since the entire reach
would have had to have an increase in upwelling and an increase of great enough
magnitude to nearly eliminate oxygenated water from entering the sediments. This would
not likely occur in one year, if at all. The reach was dominated by sandy sediments both

years suggesting that the reach was likely a depositional area with low permeability.

42

Dole-Oliver (1998) suggested that exchange processes where distinct pool-rifﬂe
sequences are not present are controlled by channel bed porosity. Sediment data
suggested that mean sediment size did not change. However, stream discharge and
stream water velocity was lower during the 1998 experiment than it was during the 1997
experiment. On 13 September 1997 stream discharge was 0.82 m3/s and mean stream
water velocity was 0.20 m/s. On 3 September 1998 stream discharge was 0.59 m3/s and
stream water velocity was 0.15 m/s. The 3 September 1998 discharge and velocity was
much lower than it was on 13 September 1997 but I have no evidence Showing that the
mean annual discharge or velocity at this site was actually lower in 1998. Another
possible explanation involves construction of a span bridge only ﬁve meters downstream
of the downstream station. Again, I have no evidence or reason to believe that
construction of the bridge impacted the study area. There is also the possibility that the
depth of the hyporheic zone in this reach varies with discharge to the extent that a
decrease of 0.23 m3/s nearly eliminated the hyporheic zone. It is difﬁcult to say how
realistic this explanation may be. It is not only realistic but probable that a change in
discharge of this magnitude could reduce the depth of the hyporheic zone several
centimeters. Unfortunately I can not be sure how deep the hyporheic zone was in 1997
because the upstream-downstream CR24 fell within the standard error of the mean
hyporheic CR24. Therefore, the depth of the hyporheic zone in 1997 was 20 cm or less.
In 1998 the depth of the hyporheic zone must have decreased anywhere from a few to 20
cm from 1997.

An alternative explanation for the reduction in upstream-downstream CR24 is that

the hyporheic zone did not decrease in size, but the microbial community within the zone

43

was reduced. Young and Huryn (1996) found that discharge within a year can
dramatically impact metabolism. Uehlinger and Naegeli (1998) observed bed-moving
spates reducing both GCP and CR24. GCP was much more impacted by the spate but
recovered faster than CR24. It is possible that a spate occurred prior to the run reducing
GCP and CR2,” only GCP already recovered and CR2, had not. This explanation is

unlikely since the run took place following a month of very limited rainfall.

Gross Community Productivity

Measures of GCP were also very similar from 1997 to 1998 at the July Augusta
Creek site (Figure 14). In 1997 and 1998 GCP was non-detectable using the hyporheic
chamber. This was actually determined a-priori with preliminary interstitial water
velocity measurements. These measurements showed that the combination of chamber
length and high velocity of interstitial water ﬂow would demand D.O. probes that could
detect changes at the 0.01 mg/l level. My probes did not have that capability. The
chamber length was arbitrarily chosen for easy transport and handling and could easily be
made longer when sensitive probes are not available. The upstream-downstream
technique measured slightly but signiﬁcantly higher GCP than the closed benthic
chambers. The mean for the two years using the upstream-downstream method was 4.8
mg 02 m'2 day" and the mean using the benthic chamber was 1.3 mg 02 m'2 day". There
are several possible explanations for the inconsistency between methods. First, the
upstream-downstream method may have included macrophyte primary production, but
this is unlikely because no macrophytes were observed in that reach. Next, there may

have been patches of high algal productivity with the reach caused by reduced ﬂow

44

9 02 m‘2 Glav'1

Gross Community Productivity

(two-station calculations)

 

 

 

 

 

 

6
O ..
5 - ........................................................................................
4 - ................. .... .............................. . .................. .<.> .................
b
3 - ............................................ 1 ............................................
2 _ .........................................................................................
§ .1.
1 - ......................................................................................
‘ O
o g Q
o - ..........................................................................................
NDND
1 e a e a a e a e a
«W9 «919 . «919” 6999 OQiOQ' .3246”
\

Figure 14. Gross community productivity was estimated using the hyporheic chamber,
benthic chamber, and the upstream-downstream technique. Both the July Augusta Creek
site and the September Kellogg Forest Augusta Creek site were monitored in 1997 and
1998. The upstream-downstream estimates were made using two-station calculations.

45

velocities or groundwater upwelling. However, there was no visible heterogeneity of
primary production. Finally, a nutrient may have been periodically limiting primary
productivity in the closed benthic chamber. The chambers were ﬂushed when dissolved
oxygen concentrations dropped below 5.0 mg 02 m‘2 clay'I or rose above 11.0 mg 02 m'2
day". While ﬂushing will decrease the chance of a nutrient becoming limiting, I cannot
be sure that a nutrient never became limiting during the experiment.

GCP measurements made at the September Kellogg Forest site were variable from
1997 to 1998 using the upstream-downstream and hyporheic chamber methods. Closed
benthic chambers produced nearly identical GCP values in 1997 and 1998. When
replicates from the two years were combined, there was no signiﬁcant difference between
the upstream-downstream method and the other two techniques (Figure 15). There was a
small but signiﬁcant difference between the hyporheic and benthic chambers with the
hyporheic measurement slightly greater. The upstream-downstream technique produced a
mean from both years of 2.3 mg 02 m'2 day", the hyporheic chamber produced a GCP of
2.6 mg 02 m'2 day" and the benthic chambers measured 0.5 mg 02 m'2 day". Again,
nutrient limitation occasionally occurring in the closed chambers seems to be the most
likely explanation for the difference between techniques. One may argue that since
benthic chambers produced the most repeatable results, as indicated by low standard
error, that these results are the most accurate. In actuality, the relatively small standard

error is a result of GCP being so near zero.

46

Combined 97198 Estimates
Of Gross Community Productivity

 

 

 

 

 

 

 

6
a
5} ...................................................................
4- ......................................................................................
‘7
g 3- ...................................................... a ..............................
'O
at I e
E 2- ....................................... a ............................................
N
O 1- .................................. i ............................................. a...
R l
'
o_ ..........................................................................................
ND
pf pf 49°” ,ef ,ef ,ef”
«9 (39 $39 99 0°99 6519
r90 599 e 690 oat 90“

Figure 15. The replicates from 1997 and 1998 for each method were combined and
compared statistically using a Mann-Whimey U test (p < 0.05). Since the upstream-
downstream technique can not be replicated, n equals two for both sites. Alike letters
indicate signiﬁcant (p < 0.05) differences.

47

P/R values

P/R values are the ratio of GCP to CR24. Within-method P/R values were Similar
between the 1997 and 1998 experiments with the exception of the upstream-downstream
method in September at Kellogg Forest (Figure 16). I combined the replicates from each
year keeping in mind that the extremely large variance associated with the upstream-
downstream method in September was due to the 1998 experiment (Figure 17).

P/R did not apply to the hyporheic chamber at the July Augusta Creek site because
GCP was non-detectable. However, NCP and NDM measurements using the hyporheic
chamber suggested that P/R would have been very similar to upstream-downstream P/R.
The mean P/R value using the upstream-downstream technique was 0.13, extremely
heterotrophic, the mean benthic chamber value was 0.94, nearly autotrophic.

Upstream-downstream P/R was not signiﬁcantly different from the other two
methods at the Kellogg Forest site due to the extremely large variance associated with this
mean. The 1997 upstream-downstream P/R value was 0.03 while the 1998 value was
108. The latter value is extremely high and obviously unrealistic. The value was
probably the result of a combination of factors. The upstream-downstream method
measured GCP higher than the other two methods in 1998. The technique may have
measured macrophyte production missed by the other two methods. Small patches of
Nasturtium oﬁ‘icinale (watercress) were observed along the margins of the reach. On the
other hand, the upstream-downstream method measured the lowest CR2, at this site in
1998. Fine sediment found in this reach resulted in a lower hydraulic conductivity. One
chamber designated for just this purpose was used to measure velocity through the

interstitial spaces. In 1998 interstitial velocity was 0.0004 rn/s, a value identical to the

48

P/R

PIR values

(two-station calculations)

 

108 o

\\

 

 

 

 

 

 

o
. O
0 . o a
«@062, «if 9% «619% SW” c5903, 993$
12° 9%“ e°s° FAQ" 0%“ e°e°
‘2‘ ‘2‘ 0909’ QQQQQQ ‘2‘ ‘2‘ QQ’QQ’ QQQQQ°
o 1997
o 1998

Figure 16. P/R values were estimated using the hyporheic chamber, benthic chamber, and
the upstream-downstream technique. Both the July Augusta Creek site and the
September Kellogg Forest Augusta Creek site were monitored in 1997 and 1998. The
upstream-downstream estimates were made using two-station calculations.

49

Combined 97l98 Estimates Of

 

 

 

 

 

 

 

PI R val u es
108 --
/’ )V
g 2 // a //
O.
a
1 i L
a
a C
0 C
«9% «9% «9% 99% 09% 99%
a“? «9 «if «if «9 «9
019 011° £39 one 091°" 01°
45° 59 86° 8° 599 99°

Figure 17. The replicates from 1997 and 1998 for each method were combined and
compared statistically using a Mann-Whitney U test. Since the upstream-downstream
technique can not be replicated, n equals two for both sites. Alike letters indicate

signiﬁcant (p < 0.05) differences.

50

velocity measured in 1997. Although exchange was not measured, a combination of
lower discharge and low hydraulic conductivity at this site may have reduced exchange
between surface and hyporheic water within the reach. This does not explain whycthe
upstream-downstream method produced a CR24 value lower than the closed benthic
chamber. CR2, measured with the benthic chamber was 0.48 02 m'2 day'1 and the
upstream-downstream method yielded an estimate of only 0.04 02 m'2 day". Perhaps this
difference was due to sampling error. McCutchan et al. (1998) estimated that respiration
values obtained using the upstream-downstream method in a low gradient stream with a
low metabolic rate would be plus or minus 34% of the actual rate. They also pointed out
that the precision and resolution of measurement used in their simulations would rarely
be achieved in the ﬁeld. Unfortunately, solid conclusions can not be drawn without
replication.

The hyporheic chamber produced a signiﬁcantly lower P/R estimate than the
benthic chamber. The benthic chamber predicted autotrophy with a P/R value of 1.36
while the hyporheic chamber predicted heterotrophy with a P/R value of 0.31.

These data compare very well to previous metabolism measurements made at
Augusta Creek. In 1973 through 1975, King and Cummins (1989) measured metabolism
using closed benthic chambers at two sites, one approximately one kilometer upstream of
the July Augusta Creek site and the other approximately one kilometer downstream of the
Kellogg Forest site. Their CR24 estimates ranged from 0.24 to 0.94 mg 02 m'2 day‘l near
the July Augusta Creek site and 0.32 to 1.41 mg 02 rn’2 day'l near the Kellogg Forest site.

GCP ranged from -0.01 to 0.59 mg 02 m“2 day" near the July Augusta Creek site and 0.53

to 2.16 mg 02 m'2 day" near the Kellogg Forest site. Bott et al. (1985) also made seasonal

51

metabolism measurements on Augusta creek using closed benthic chambers in 1976.
Their CR24 values ranged from 1.53 to 2.66 mg 02 m'2 day" approximately one kilometer
downstream of the July Augusta Creek site and 1.55 to 2.97 mg 02 m'2 day"
approximately one kilometer downstream of the Kellogg Forest site. GCP ranged from
0.50 to 2.80 mg 02 rn’2 day" and from 0.72 to 3.40 mg 02 m'2 day" at the two sites
respectively. My measurements made using the closed benthic chambers fall very near if

not within these ranges.

52

CONCLUSIONS

It is difﬁcult to predict the amount of error associated with the upstream-
downstrearn method since it cannot be replicated. However, the method produced
comparable results at the two sites from 1997 to 1998. The methods used for these
experiments involving conservative and volatile tracers to estimate oxygen diffusion are
quite involved. Experience with these techniques could reduce sampling error. Very
specialized equipment along with a trained operator are necessary to conduct these
measurements. With experience, the technique can give reliable results. Unfortunately,
the upstream-downstream technique without the conservative and volatile tracers is often
considered suspect due to the uncertainty associated with reaeration estimates derived
from empirical relations in the literature. When parameters from the two-station
experiments were calculated using a single station technique, they gave very different
results in an unpredictable direction. For example, CR2, values calculated using a single
station were lower at the July Augusta Creek site for both years and higher at the Kellogg
Forest. GCP was lower than the two-station technique in 1997 at the July Augusta Creek
site, and the single-station was much higher than the two-station technique at the same
site in 1998. Finally, variance between years tended to increase when the single-station
calculations were used in place of the two-station calculation for all of the parameters.

The simple design of the hyporheic chambers should reduce sampling error.
However, the variance associated with these chamber measurements was often higher
than that associated with closed benthic chambers. This error might have been reduced
substantially if interstitial water velocity had been measured in individual replicates and

applied to those speciﬁc data. Velocity measurements involved spiking a randomly

53

chosen chamber with concentrated sodium chloride. This was not done in each chamber
because sediments from the replicates were taken to the lab and analyzed for chlorophyll
a and organic content. Concentrated sodium chloride could have had deleterious effects
on these data and metabolism measurements. Ideally, velocity measurements Should be
made after metabolism measurement in each replicate chamber. Preliminary studies
suggested that using one chamber for velocity would provide a good estimate of velocity
for all of the chambers, perhaps because measurements were made at a small scale in a
single riffle. However, the hyporheic zone has been shown to be extremely variable and
the magnitude of variability is likely stream and even riflle dependent.

Closed benthic chambers maintained low variability between replicates for the
entire study, making them extremely good candidates for comparative studies of surface
sediment metabolism. Even so, special attention should be given to factors limiting
primary productivity, especially if comparisons are carried out between streams, as the
time involved for a nutrient to become limiting would be expected to vary widely from
stream to stream.

The three methods used simultaneously can reveal error otherwise overlooked
while compartmentalizing metabolism and therefore provide much more information
about the system. Bott et a1. (1997) cautioned that one should not collect metabolism
data in a routine manner without assessing potential sources of error. Use of the benthic
chamber provides estimates of surface sediment metabolism. Combining these estimates
with estimates from hyporheic chambers would allow surﬁcial sediment respiration
associated with autotrophs and heterotrophs to be subtracted from hyporheic chamber

measurements, providing estimates of hyporheic metabolism to a known depth. This

54

value in turn could be used in conjunction with the whole system estimate of the
upstream-downstream method to provide an estimate of mean hyporheic depth for a given

reach of stream.

55

APPENDICES

56

APPENDIX A

Substrate data based on three cores taken from hyporheic chambers.

Sample Date Chlorophyll a Percent organic
(Phaeophytin corrected mg m‘z) content to 20 cm

depth

AuSable R. 6/96 ---- 0.9 $0.1
N.B. AuSable 6/96 105.3 :1: 24.5 2.9 :1: 1.2
N.B. AuSable 6/97 145.1 :1: 47.3 2.6 :1: 1.7

Jordan 9/96 736.6 :1: 161.2 1.8 $1.3

Red Cedar 6/97 55.4 :t 9.7 1.0 a: 0.0

Augusta Cr. 7/97 85.4 :1: 7.4 2.9 :1: 1.0

Augusta Cr. 7/98 102.8 :t 26.7 1.1 i 0.4

Augusta Cr. 9/97 169.5 1 21.9 1.2 d: 0.2

(Kellogg Forest)
Augusta Cr. 9/98 49.5 :h 21.9 0.9 i 0.1

 

(Kellogg Forest)

 

57

Percent
inorganic
> 8 mm

42.2 :t 8.1
04i01
<10*
6L8i61
75.8 :1: 9.9
19.4 :1: 3.2
120i35

Percent

inorganic 2.0 -

8.0 mm

18.2 :1: 1.2
25:01
<10*
19.2 :h 2.8
11.4 i 4.5
21.5 i 3.3
29.0 :1: 0.4

Percent
inorganic

0.250 - 2.0 mm

388i78
946i02
<10*
177i34
1zoisl
514:29
503:33

Percent

inorganic
0.125 -0.250

mm

0.5 :1: 0.3
2.4 d: 0.1
<10*
05i01
0.4 i 0.2
1.4 i 0.3
7Ai06

 

Percent

inorganic <
0.125 mm

 

OZiOJ
01:00
>90*
OJiOD
03i02
02100
12:02

 

* Estimated value-the sample would not pass through a sieve due to consolidated clay.

APPENDIX B

Example of calculations for each method using data from Augusta Creek on 29 July 1997.

Hyporheic Chamber
Discharge: 1.224 m3 hr‘l Area: 0.153 m3 Interstitial-water/sediment area: 0.35
Interstitial discharge: 0.4284 m3 hr"

DO change from upstream to downstream port (mg/l) Rate of change (g 0, 111‘2 hr")

59

Time Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3
00:00 -0.4 -0.3 -0.1 -l . 12 —0.84 -0.28
01 :00
02:00 -0.5 -0.1 -0.55 -1.40 -0.28 -1.54
03:00
04:00 -0.5 0 -0.5 -1.40 0.00 -1.40
05:00
06:00 -0.6 -0.1 -0.35 -1.68 -0.28 -0.98
07:00 -0.4 -0.3 -0.3 -l.12 -0.84 -0.84
08:00 -0.5 -0.2 -0.3 -1.40 -0.56 -0.84
09:00 -0.4 -0.2 -0.3 -l .12 -0.56 -0.84
10:00 -0.5 -0.4 -0.25 -l .40 -1.12 —0.70
11:00 -0.6 -0.3 -0.25 -1.68 -0.84 -0.70
12:00 -0.55 -0.2 -0.4 -l.54 —0.56 -l.12
13:00 -0.55 -0.2 -0.4 -1.54 -0.56 -1.12
14:00 -0.55 -0.3 -0.4 -l.54 -0.84 -1.12
15:00 -0.65 -0.3 -0.4 -1 .82 -0.84 -1 . 12
16:00 -0.65 -0.2 -0.3 -1.82 -0.56 -0.84
17:00 -0.65 -0.4 -0.3 -1.82 -1.12 -0.84
18:00 -0.45 -0.3 -0.1 -1.26 -0.84 -0.28
19:00 -0.4 -0.5 -0.1 -1.12 -1.40 -0.28
20:00 -0.4 -0.4 -0.25 -1.12 -1.12 -0.70
21:00 -0.4 -0.4 -0.25 -1.12 -1.12 -0.70
22:00 -0.4 -0.3 -0.15 -l .12 -0.84 -0.42
23:00 -0.4 -0.4 -0.1 -1.12 -1.12 -0.28

# of light hours 14.58
# of night hours 9.42

Rep 1 Rep 2 Rep 3
# of light readings 15 15 15
# of night readings 6 6 6
NCP (g Oz/mz‘dKY) -20.82 -12.52 -1 1.70
CR (8 Oz/mz‘hr) 1.39 0.77 0.81

GCP (g Oz/mz‘day) ND ND ND
NDM (g 02/m2*dayl -33.44 -18.56 -19.36
P/R 0.00 0.00 0.00

Benthic Chamber

Chamber volume (1): 9.5, 9.5, 9.5 Sediment conversion to m’: 15

DO change from previous reading (mg/I) Rate of change (g 02 m" hr")

Time Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3
00:00 -0.2 -0.3 -0.4 -0.03 -0.04 -0.06
01:00 -0.25 -0.3 -0.45 -0.04 -0.04 -0.06
02:00 -0.25 -0.3 -0.45 -0.04 -0.04 -0.06
03:00 -0.3 -0.25 -0.35 -0.04 -0.04 -0.05
04:00 -0.3 -0.25 -0.35 -0.04 -0.04 -0.05
05:00 -0.4 -0.45 -0.7 -0.06 -0.06 -0.10
06:00 -0.4 -0.45 -0.7 -0.06 -0.06 -0.10
07:00 -0.2 -0.2 -0.4 -0.03 -0.03 -0.06
08:00 -0.1 -O.1 -0.1 -0.01 -0.01 -0.01
09:00 -0.1 0 -0.1 -0.01 0.00 -0.01
10:00 0.3 0.3 0 0.04 0.04 0.00
11:00 0.5 1 0.3 0.07 0.14 0.04
12:00 1.1 0.5 1.9 0.16 0.07 0.27
13:00 0.5 0.5 0.4 0.07 0.07 0.06
14:00 -0.1 0.2 0.2 -0.01 0.03 0.03
15:00 0.1 0 0 0.01 0.00 0.00
16:00 -0.2 -0.2 -0.3 -0.03 -0.03 -0.04
17:00 -0.2 -0.1 -0.3 -0.03 -0.01 -0.04
18:00 -0.3 04 -0.3 -0.04 -0.06 -0.04
19:00 -0.2 -0.3 -0.3 -0.03 -0.04 -0.04
20:00 -0.3 -0.3 -0.4 -0.04 -0.04 -0.06
21 :00 —0.4 -0.4 -0.5 -0.06 -0.06 -0.07
22:00 -0.3 -0.3 -0.6 -0.04 -0.04 -0.09
23:00 -0.3 -0.4 -0.4 -0.04 -0.06 -0.06

60

# of light hours 14.58
# of night hours 9.42
Rep 1 Rep 2 Rep 3
# of light readings 15 15 15
# of night readings 9 9 9
NCP (g Oz/mz'fday) 0.06 0.07 0.01
CR (g Ozlmuhr) 0.04 0.05 0.07
GCP (g Oz/m2*day) 0.70 0.78 1.06
NDM (g Oz/m2*day) -0.33 -0.36 -0.61
P/R 0.68 0.69 0.63
Upstream-downstream
koxygen * travel time = 0.1140*l.625 = 0.18525
(two station) (one station)
Time Temp 1 Temp 2 Temp upstream downstreamChange DO Mean DO
AM 11:00 19.00 19.00 19.00 7.70 7.60 -0.10 7.65
PM 12:00 19.50 19.50 19.50 7.80 7.60 -0.20 7.70
1:00 20.00 20.20 20.10 7.90 7.70 -0.20 7.80
2:00 21.20 21.30 21.25 7.90 7.90 0.00 7.90
3:00 21.50 21.50 21.50 7.70 7.60 -0.10 7.65
4:00 22.20 22.50 22.35 7.70 7.60 -0.10 7.65
5:00 22.50 22.70 22.60 7.50 7.40 -0.10 7.45
6:00 22.50 22.80 22.65 7.30 7.20 -0.10 7.25
7:00 22.20 22.50 22.35 7.00 7.00 0.00 7.00
8:00 22.00 22.20 22.10 6.80 6.70 -0.10 6.75
9:00 21.50 22.00 21.75 6.60 6.50 -0.10 6.55
10:00 21.20 21.50 21.35 6.60 6.60 0.00 6.60
11:00 20.50 21.00 20.75 6.60 6.50 -0.10 6.55
AM 12:00 20.00 20.50 20.25 6.70 6.60 -0.10 6.65
1:00
2:00 19.00 19.20 19.10 6.80 6.70 -0.10 6.75
3:00
4:00 18.20 18.70 18.45 7.00 6.90 -0.10 6.95
5:00
6:00 17.00 17.50 17.25 7.30 7.20 -0.10 7.25

61

(at 841 ft)

100 % sat
9.00
8.91
8.80
8.60
8.57
8.40
8.38
8.38
8.40
8.46
8.53
8.59
8.69
8.78

8.98

9.10

9.33
9.44
9.51
9.51
9.45
9.38

7:00
8:00
9:00
10:00
11:00

DO deﬁcit

1.35
1.21
1.00
0.70
0.92
0.75
0.93
1.13
1.40
1.71
1.98
1.99
2.14
2.13

2.23

2.15

2.08
1.94
1.86
1.71
1.55
1.33

16.50
16.20
16.20
16.50
17.00

16.90
16.50
16.50
16.80
17.00

0.25
0.22
0.18
0.13
0.17
0.14
0.17
0.21
0.26
0.32
0.37
0.37
0.40
0.39

0.41

0.40

0.39
0.36
0.34
0.32
0.29
0.25

16.70
16.35
16.35
16.65
17.00

62

7.60
7.70
7.90
8.00
8.20

(100%-mean) (instantaneous) (Flux into water)

-0.25
-0.22
-0. l 8
-0.13
-0.17
-0.14
-0.17
-0.21
-0.26
-0.32
-0.37
-0.37
-0.40
-0.39

-0.41

-0.40

-0.39
-0.36
-0.34
-0.32
-0.29
-0.25

7.40
7.60
7.70
7.80
7.90

(two station)
(diff between SID’S) (difference per hr)
Reaeration Flux -Reaeration FluxCorr. rate of chg Corr. rate of chg

-0.35
-0.42
-0.38
-0.13
-0.27
-0.24
-0.27
-0.31
-0.26
-0.42
-0.47
-0.37
-0.50
-0.49

-0.51

-0.50

-0.49
-0.56
-0.44
-0.52
-0.49
-0.55

-0.20
-0.10
-0.20
-0.20
-0.30

7.50
7.65
7.80
7.90
8.05

(one station)

0.10
-0. 17
-0.08
-0.03
-0.42
-0.14
-0.37
-0.41
-0.51
-0.57
-0.57
-0.32
-0.45
-0.29

-0.36

-0.30

-0.24
-0.11
—0.19
-0.17
-0.19
-0.10

Single-station calculations

# of light hours
# of night hours

# of light readings
# of night readings
Rep 1

NCP (g Oz/m2*day)
CR (g 02/m2*hr)
GCP (g 02/m2*day)
NDM (g Oz/m2*day)
P/R

Two-station calculations

# of light hours
# of night hours

# of light readings
# of night readings
Rep 1

NCP (g 02/m2*day)
CR (g 02/m2*hr)
GCP (g 02/m2*day)
NDM (g 02/m2*day)
P/R

63

14.58
9.42

-12.01
1.00
2.59

-21.44
0.11

14.58
9.42

15

-17.08
1.46
4.23

-30.85
0.12

LITERATURE CITED

LITERATURE CITED

Bott, T.L. 1996. Primary productivity and community respiration. In Hauer, F .R. and
G.A. Lamberti. 1996. Methods in stream ecology. Academic Press, Inc., London.
533-556.

Bott, T.L., J .T. Brock, A. Baattrup-Pedersen, P.A. Chambers, W.K. Dodds, K.T.
Himbeault, J .R. Lawrence, D. Planas, E. Snyder, and GM. Wolfaardt. (1997). An
evaluation of techniques for measuring periphyton metabolism in chambers.
Canadian Journal of Fisheries and Aquatic Science 54:715-725.

Bott, T.L., J .T. Brock, C.E. Cushing, S.V. Gregory, D. King, and RC. Petersen. 1978.
A comparison of methods for measuring primary productivity and community
respiration in streams. Hydrobiologia 6023-12.

Bott, T.L., J .T. Brock, C.S. Dunn, R.J. Naiman, R.W. Ovink, and RC. Petersen. 1985.
Benthic community metabolism in four temperate stream systems: An inter-biome
comparison and evaluation of the river continuum concept. Hydrobiologia 123:3-
45.

Dole-Olivier, M.J. 1998. Surface water-groundwater exchanges in three dimensions on a
blackwater of the Rhone River. Freshwater Biology 40193-1 09.

Fetter, CW. 1994. Applied Hydrogeology. 3rd edition. Macmillan College Publishing
Company, New York. 691 pp.

Ford TE. and R.J. Naiman. 1989. Groundwater-surface water relationships in boreal
forest watersheds: dissolved organic carbon and inorganic nutrient dynamics.
Canadian Journal of Fisheries and Aquatic Science 46:41-49.

Fuss, C.L., and LA. Smock. 1996. Spatial and temporal variation of microbial
respiration rates in a blackwater stream. Freshwater Biology 36:339-349.

Grimm, N.B. and 8.6. Fisher. 1984. Exchange between interstitial and surface water:
implications for stream metabolism and nutrient cycling. Hydrobiologia 111:219-
228.

Hauer, RR. and G.A. Lamberti. 1996. Methods in stream ecology. Academic Press, Inc.,
London. 674 pp.

Hendricks, SP. and D.S. White. 1991. Physicochemical patterns within a hyporheic zone
of northern Michigan river, with comments on surface water patterns.

Canadian Journal of Fisheries and Aquatic Science 48:1645-1654.

Hendricks, SP. and D.S. White. 1995. Seasonal biogeochemical patterns in surface

65

water, subsurface water, subsurface hyporheic, and riparian ground water in a
temperate stream ecosystem. Archiv ﬁjr Hydrobiologie 134(4):459-490.

Homer, RR, and EB. Welch. 1981. Stream periphyton development in relation to
current velocity and nutrients. Canadian Journal of Fisheries and Aquatic Science
38:449-457.

Homer, R.R., E.B. Welch, M.R. Seeley, and J .M. Jacoby. 1990. Response of periphyton
to changes in current velocity, suspended sediment and phosphorus concentration.
Freshwater Biology 24:215-232.

Jones, J .B., S.G. Fisher, and NB. Grimm. 1995. Vertical hydrologic exchange and
ecosystem metabolism in a Sonoran Desert stream. Ecology 76(3):942-952.

Jones, J .B., S.G. Fisher, and NB. Grimm. 1995. Nitriﬁcation in the hyporheic zone of a
desert stream ecosystem. Journal of the North American Benthological Society
14(2):249-25 8.

King, D.K. and K.W. Cummins. 1989. Autotrophic-heterotrophic community
metabolism relationships of a woodland stream. Journal of Freshwater Ecology.
5(2):205-218.

Marzolf, E.R., P.J. Mulholland, and AD. Steinman. 1994. Improvements to the diurnal
upstream-downstream dissolved oxygen change technique for determining whole-
stream metabolism in small streams. Canadian Journal of Fisheries and Aquatic
Science 51:1591-1599.

Marzolf, E.R., P.J. Mulholland, and AD. Steinman. 1998. Reply: Improvements to the
diurnal upstream-downstream dissolved oxygen change technique for determining
whole-stream metabolism in small streams. Canadian Journal of Fisheries and
Aquatic Science 55: 1786-1787.

McCutchan, J .H., W.M. Lewis, and J .F . Saunders. 1998. Uncertainty in the estimation of
stream metabolism from open-channel oxygen concentrations. Journal of the
North American Benthological Society 17(2): 155-164.

McDiffett, W.F., A.E. Carr, and D.L. Young. 1972. An estimate of primary productivity
in a Pennsylvania trout stream using a diurnal oxygen curve technique. American
Midland Naturalist 87(2):564-570.

McIntire, C.D., R.L. Garrison, H.K. Phinney, and CE. Warren. 1964. Primary production
in laboratory streams. Limnology and Oceanography 9:92-102.

Mulholland, P.J., E.R. Marzolf, J .R. Webster, D.R. Hart, and SP. Hendricks. 1997.
Evidence that hyporheic zones increase heterotrophic metabolism and phosphorus
uptake in forest streams. Limnology and Oceanography 42(3):443-451.

66

Naegeli, M.W., U. Hartmann, E.I. Meyer, and U. Uehlinger. 1995. POM-dynamics and
community respiration in the sediments of a ﬂoodprone prealpine river (N ecker,
Switzerland). Archiv ﬁir Hydrobiologie 133(3):339-347.

Odum, HT. (1956) Primary production in ﬂowing waters. Limnology and Oceanography
1:102-117.

Pusch M. and J. Schwoerbel. 1994. Community respiration in hyporheic sediments of a
mountain stream (Steina, Black Forest). Archiv ﬁir Hydrobiologie 130(1):35-52.

Rathbun, R.E., D.W. Stephens, D.J. Schultz, and D.Y. Tai. 1978. Laboratory studies of
gas tracers for reaeration. Proc. ASCE 104:215-229.

Rutherford, J .E. and H.B.N. Hynes. 1987. Dissolved organic carbon in streams and
groundwater. Hydrobiologia 154:33-48.

Stanford, J .R. and J .V. Ward. 1993. An ecosystem perspective of alluvial rivers:
connectivity and the hyporheic corridor. Journal of the North American
Benthological Society 12:48-60.

Triska, F .J ., V.C. Kennedy, R.J. Avanzino, G.W. Zellweger, and KB Bencala. 1989.
Retention and transport of nutrients in a third-order stream in Northwestern
California: Hyporheic processes. Ecology 70(6): 1 893-1905.

Triska, F.J., V.C. Kennedy, R.J. Avanzino, G.W. Zellweger, and KB. Bencala. 1989.
Retention and transport of nutrients in a third-order stream: channel processes.
Ecology 70(6): 1877-1892.

Triska, F.J., V.C. Kennedy, R.J. Avanzino, G.W. Zellweger, and KB. Bencala. 1990. In-
situ retention-transport response to nitrate loading and storm discharge in a third-
order stream. Journal of the North American Benthological Society 9:229-239.

Uehlinger, U. and M.W. Naegeli. 1998. Ecosystem metabolism, disturbance, and
stability in a prealpine gravel bed river. Journal of the North American
Benthological Society 17(2):165-178.

Valett, H.M., S.G. Fisher, and EH. Stanley. 1990. Physical and chemical characteristics
of the hyporheic zone of a Sonoran Desert Stream. Journal of the North American
Benthological Society 9:201 -2 1 5.

Vannote, R.L., G.W. Minshall, K.W. Curnmins, J .R. Sedell, and C. E. Cushing. 1980.
The river continuum concept. Canadian Journal of Fisheries and Aquatic Science
37:130-137.

Young, R.G., and AD. Huryn. 1998. Comment: Improvements to the diurnal upstream-
downstream dissolved oxygen change technique for determining whole-stream

67

metabolism in small streams. Canadian Journal of Fisheries and Aquatic Science
55:1784-1785.

Young, R.G., and AD. Huryn. 1996. Inter-annual variation in discharge controls
ecosystem metabolism along a grassland river continutun. Canadian Journal of
Fisheries and Aquatic Science 53:2199-2211.

68

HICHIan STATE UNIV. LIBRQRIES
111|111111111111111111111111111111111111111
31293018129431