Whole-stream metabolism in two montane streams: Contribution of the hyporheic zone
ABSTRACT We used whole-stream and benthic chamber methods to measure rates of metabolism and determine the contribution of the hyporheic zone to ecosystem respiration (R) in two streams with differing surface–subsurface exchange characteristics, Rio Calaveras and Gallina Creek, New Mexico. We used the difference between whole-stream and benthic R to calculate the rate of hyporheic zone R and coupled this estimate to an independent measure of hyporheic sediment R to estimate the cross-sectional area of the hyporheic zone (AH) for two reaches from each stream. Conservative tracer injections and solute transport modeling were used to characterize surface–subsurface hydrologic exchange by determining values of the cross-sectional area of the transient storage zone (As). The hyporheic zone contributed a substantial proportion of whole-stream R in all four study reaches, ranging from 40 to 93%. Wholestream R, hyporheic R, and percent contribution of hyporheic R all increased as transient storage increased, with whole-stream and hyporheic R exhibiting significant relationships with As. All three measures of respiration and values of AH were much greater for both reaches of the stream with greater surface–subsurface exchange. AH is valuable for cross-site comparisons because it accounts for differences in rates of both benthic and hyporheic sediment R and can be used to predict the importance of the hyporheic zone to other stream ecosystem processes. Yes Yes
- SourceAvailable from: onlinelibrary.wiley.com[Show abstract] [Hide abstract]
ABSTRACT: Biogeochemical cycling within river ecosystems is strongly influenced by geomorphic and hydrologic dynamics. To scale point observations of temperature and dissolved oxygen (DO) to a hydrologically complex and dynamic three-dimensional river-floodplain-aquifer system, we integrated empirical models of temperature and biotic oxygen utilization with a recently published hydrogeomorphic model. The hydrogeomorphic model simulates channel flow, floodplain inundation, and surface-subsurface water exchange on the 16 km2 Nyack Floodplain, Middle Fork Flathead River, Montana, USA. Model results were compared to observed data sets of DO to test the hypothesis that complexity in spatiotemporal patterns of biogeochemistry emerges from a comparatively simple representation of biogeochemical processes operating within a multidimensional hydrologic system. The model explained 58% of the variance in 820 DO measurements that spanned the study site longitudinally, laterally, vertically, and across river discharge conditions and seasons. We also used model results to illustrate spatial and temporal trends of temperature and DO dynamics within the shallow alluvial aquifer, which is an extensive hyporheic zone because subsurface alluvial flow paths are recharged primarily by channel water. Our results underscore the importance of geomorphic, hydrologic, and temperature dynamics in driving river ecosystem processes, and they demonstrate how a realistic representation of a river's physical template, combined with simple biogeochemical models, can explain complex patterns of solute availability.Journal of Geophysical Research (Biogeosciences). 12/2012; 117(G4).
- [Show abstract] [Hide abstract]
ABSTRACT: An extension of classical harmonic analysis is used to determine the Fourier coefficients with a constraint of forcing the function through a specified value for developing an extended stochastic harmonic analysis (ESHA). The methodology was applied to the diurnal variations of dissolved oxygen (DO) in streams within a semiempirical modeling scheme to demonstrate the formulation of an ESHA algorithm. Since DO values are often measured at different times during the day, the ESHA algorithm was used to convert data measured at any clock time to those at a standard or reference time. The model was evaluated for five different streams in Minnesota that capture the impact of different ecoregions and different sizes of watersheds. Data were normalized to increase the general applicability of the fitted parameters. The ESHA was successful in representing observed diurnal variations in DO. The root-mean-square error (RMSE) for predicting hourly DO and standard DO ranged respectively from 0.43 to 0.77 mg/L and 0.37 to 0.90 mg/L among the five streams. Estimated model parameters were robust in terms of both spatial and temporal variations. Parameter uncertainties and associated model sensitivity were demonstrated analytically. The developed algorithm can be used for similar applications in numerous science and engineering disciplines. In particular, it is a potentially useful tool for total maximum daily load (TMDL) assessment of aquatic ecosystem health across a range of spatial and temporal scales.Water Resources Research 01/2007; 43(8). · 3.15 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Implementing stream restoration techniques and best management practices to reduce nonpoint source nutrients implies enhancement of the assimilative capacity for the stream system. In this paper, a Bayesian method for evaluating this component of a total maximum daily load (TMDL) load capacity is developed and applied. The joint distribution of nutrient retention metrics from a literature review of 495 measurements was used for Monte Carlo sampling with a process transfer function for nutrient attenuation. Using the resulting histograms of nutrient retention, reference prior distributions were developed for sites in which some of the metrics contributing to the transfer function were measured. Contributing metrics for the prior include stream discharge, cross-sectional area, fraction of storage volume to free stream volume, denitrification rate constant, storage zone mass transfer rate, dispersion coefficient, and others. Confidence of compliance (CC) that any given level of nutrient retention has been achieved is also determined using this approach. The shape of the CC curve is dependent on the metrics measured and serves in part as a measure of the information provided by the metrics to predict nutrient retention. It is also a direct measurement, with a margin of safety, of the fraction of export load that can be reduced through changing retention metrics. For an impaired stream in western Oklahoma, a combination of prior information and measurement of nutrient attenuation was used to illustrate the proposed approach. This method may be considered for TMDL implementation.Water Resources Research 01/2008; 44(8). · 3.15 Impact Factor
Limnol. Oceanogr., 46(3), 2001, 523–531
? 2001, by the American Society of Limnology and Oceanography, Inc.
Whole-stream metabolism in two montane streams: Contribution of the hyporheic zone
Christine S. Fellows1, H. Maurice Valett2, and Clifford N. Dahm
University of New Mexico, Department of Biology, Albuquerque, New Mexico 87131
We used whole-stream and benthic chamber methods to measure rates of metabolism and determine the contri-
bution of the hyporheic zone to ecosystem respiration (R) in two streams with differing surface–subsurface exchange
characteristics, Rio Calaveras and Gallina Creek, New Mexico. We used the difference between whole-stream and
benthic R to calculate the rate of hyporheic zone R and coupled this estimate to an independent measure of hyporheic
sediment R to estimate the cross-sectional area of the hyporheic zone (AH) for two reaches from each stream.
Conservative tracer injections and solute transport modeling were used to characterize surface–subsurfacehydrologic
exchange by determining values of the cross-sectional area of the transient storage zone (As). The hyporheic zone
contributed a substantial proportion of whole-stream R in all four study reaches, ranging from 40 to 93%. Whole-
stream R, hyporheic R, and percent contribution of hyporheic R all increased as transient storage increased, with
whole-stream and hyporheic R exhibiting significant relationships with As. All three measures of respiration and
values of AHwere much greater for both reaches of the stream with greater surface–subsurface exchange. AHis
valuable for cross-site comparisons because it accounts for differences in rates of both benthic and hyporheic
sediment R and can be used to predict the importance of the hyporheic zone to other stream ecosystem processes.
Aquatic ecologists are increasingly aware of the impor-
tance of interactions between surface water and groundwater
to the functioning of aquatic ecosystems (Boulton et al.
1998; Winter et al. 1998; Jones and Mulholland 2000). The
region of mixing between groundwater and stream water
(i.e., the hyporheic zone, sensu Orghidan 1959; Triska et al.
1989b) influences ecosystem functioning because its sedi-
ments contain metabolically active microbial assemblages
(Grimm and Fisher 1984; Pusch and Schwoerbel 1994; Jones
et al. 1995b; Naegeli and Uehlinger 1997). Several recent
studies have focused on the role of the hyporheic zone in
the retention (Speaker et al. 1984; Triska et al. 1989b, 1990;
Hendricks and White 1991; Triska et al. 1993; Valett et al.
1996) and the transformation (Duff and Triska 1990; Jones
et al. 1995a; Holmes et al. 1996) of biologically important
solutes. Far less work has focused on the influence of the
hyporheic zone on whole ecosystem energetics (but see
1Present address: Centre for Catchment and In-Stream Research,
Faculty of Environmental Sciences, Griffith University, Nathan,
Queensland 4111, Australia (email@example.com).
2Present address: Virginia Polytechnic Institute and State Uni-
versity, Department of Biology, Blacksburg, Virginia 24061.
The authors thank everyone who helped in the field, particularly
Jim Thibault, Michelle Baker, Miranda Dendy, and Lisa Roberts.
Rob Runkel, Ken Bencala, Doug Moyer, and John Morrice gave
valuable advice on using the OTIS model. Michelle Baker ran the
hyporheic sediment incubations for Gallina Creek. Todd Royer pro-
vided equipment and helped with benthic chamber metabolism mea-
surements in 1997. Pat Mulholland provided equipment and con-
ducted whole-stream metabolism measurements for the upper reach
of Gallina Creek. Jen Tank conducted the solute injection for the
upper reach of Gallina Creek used for modeling. Two anonymous
reviewers provided helpful comments on an earlier draft. This re-
search was funded by NSF grants DEB 9420510 to H.M.V. and
M.E. Campana, and DEB 9816087 to C.N.D. and a NSF Graduate
Fellowship to C.S.F.
Grimm and Fisher 1984; Mulholland et al. 1997; Naegeli
and Uehlinger 1997).
In conceptual models presented by Findlay (1995) and
Valett et al. (1996), the contribution of the hyporheic zone
to stream ecosystem functioning depends on the types and
rates of metabolic processes occurring in the hyporheic zone,
the proportion of stream discharge routing through the hy-
porheic zone, and its impact on hydrologic residence time.
As rates of metabolism in the hyporheic zone and the vol-
ume of subsurface sediment actively exchanging water with
the stream channel both increase, the importance of the hy-
porheic zone to surface processes increases.
Studies investigating the relationship between surface–
subsurface hydrologic exchange and stream ecosystem func-
tioning have provided indirect evidence of the importance of
the hyporheic zone. Valett et al. (1996) found a strong re-
lationship between nitrate-nitrogen uptake length and ex-
change for three streams in catchments of differing lithology.
Minshall et al. (2000) found that retention of fine particulate
organic matter was correlated with hydrologic exchange
across six sites of varying size and discharge. Ecosystem
respiration (R) was greater in the stream with greater sur-
face–subsurface exchange in a comparison of two Appala-
chian streams with similar hydrological and chemical char-
acteristics (Mulholland et al. 1997). None of these studies
quantified the contribution of the hyporheic zone, however.
In contrast, Grimm and Fisher (1984) and Naegeli and Ueh-
linger (1997) quantified the contribution of the hyporheic
zone to whole-stream R, but neither study included quanti-
tative estimates of surface–subsurface exchange.
Coupled measures of metabolism and exchange provide
links between ecosystem processes and hydrology that allow
for comparisons among differing stream ecosystems. How-
ever, understanding the contribution of the hyporheic zone
to stream ecosystem processes requires not only quantitative
measures of hydrologic exchange, but also process measure-
ments made both at the whole-stream level and at scales that
will allow distinction between the surface and hyporheic
Fellows et al.
zone. In this research, we address how differences in sur-
face–subsurface exchange influence stream ecosystem me-
tabolism with a focus on the relative contribution of the hy-
porheic zone to ecosystem R. Our approach is to use the
difference between whole-stream and benthic R to determine
the rate of R in the hyporheic zone. We then couple this
estimate to an additional and independent measure of hy-
porheic R to estimate the size of the hyporheic zone. Further,
we tie rates of metabolic processes and their distribution
between surface and hyporheic subsystems to ecosystem hy-
drology by applying solute transport models to quantify fea-
tures of surface-hyporheic exchange.
The study was conducted in two headwater streams of
north-central New Mexico, under baseflow conditions during
the summers of 1996 and 1997. Previous work has shown
that these streams differ greatly in the extent of groundwa-
ter–surface water exchange, with little exchange at Rio Ca-
laveras and generally greater exchange at Gallina Creek
(Valett et al. 1996; Morrice et al. 1997).
Rio Calaveras is located in the Jemez Mountains at an
elevation of 2475 m, approximately 63 km west of Los Al-
amos, New Mexico. A 110 m reach of Rio Calaveras was
chosen for study in 1996, and in 1997 the experimental reach
was shortened to include only the lower half of the reach in
order to decrease water travel time. The two lengths of the
stream are referred to as the long and short reaches, respec-
tively. Gallina Creek is in the Sangre de Cristo Mountains
at an elevation of 2524 m, approximately 31 km northeast
of Taos, New Mexico. The study reach used in 1997 was
400 m upstream of that used in 1996. The two reaches are
referred to as the lower and upper reaches, respectively.
Solute injections—Solute injections were used to quantify
aspects of surface hydrology and surface–subsurface water
interactions within each stream reach. For a detailed descrip-
tion of these techniques, see Stream Solute Workshop
(1990). We injected a conservative tracer (Br?or Cl?) and
monitored solute concentrations at a station downstream us-
ing in-stream ion-specific electrodes (ISE, Orion) or an elec-
trical conductivity meter (YSI or VWR). Solute was injected
for 1 to 6 h, depending on the time required to generate
complete mixing throughout the reach. Concentrations of the
conservative tracer were measured directly on water samples
by ion-chromatography (Dionex DX-100) or were deter-
mined through regression analysis relating results from the
ISE probes or electrical conductivity to selected water sam-
ples analyzed by ion-chromatography. Solute injections for
modeling were conducted once at each reach in conjunction
with measures of whole-stream metabolism, generally in ear-
ly to midafternoon.
Background-corrected plateau concentrations of conser-
vative tracers were used to calculate stream discharge (Q)
for each solute injection (Triska et al. 1989a; Gordon et al.
1992). Replicate stream water samples were taken at an up-
stream station (far enough below the solute addition to allow
for complete mixing) and the downstream station to quantify
lateral inflow (Stream Solute Workshop 1990). In general, Q
was nearly constant along each of the chosen stream reaches
and mean values of Q at the upstream and downstream sta-
tions were used in solute modeling and metabolism calcu-
lations. For the upper reach of Gallina Creek, Q was cal-
culated as the average of four longitudinally distributed
sampling points. A single tracer injection was conducted at
each site in 1996, but multiple injections were performed in
1997 to resolve temporal changes in Q for more accurate
whole-stream metabolism calculations. Nominal travel time
(TN) was calculated for each reach as the time required to
reach half plateau concentration at the downstream station.
Reported values of water velocity were calculated as nomi-
nal travel time divided by reach length.
Solute curves (downstream concentration vs. time) were
used to visually fit model outputs from a one-dimensional
transport with inflow and storage model (OTIS, Runkel
1998), a mathematical simulation based on the transient stor-
age model presented by Bencala and Walters (1983). Sur-
face–subsurface water exchange can be characterized using
transient storage models that employ a hypothetical, non-
advective storage zone to account for flow paths moving
much slower than the advective velocity of the stream chan-
nel (Bencala and Walters 1983; Stream Solute Workshop
1990; D’Angelo et al. 1993; Harvey et al. 1996; Morrice et
al. 1997; Runkel 1998). The ratio between the cross-section-
al area of this storage zone (As) and of the above-ground
stream channel (A) normalizes the extent of transient storage
to stream size and has served as a useful comparative mea-
sure of hyporheic zone size (D’Angelo et al. 1993; Valett et
al. 1996; Morrice et al. 1997; Mulholland et al. 1997). The
parameters estimated through visual fitting were stream
cross-sectional area (A, m2), storage zone cross-sectional
area (As, m2), dispersion (D, m2s?1), and the storage zone
exchange coefficient (?, s?1). Additionally, after visual best
fits were obtained, parameter estimates were entered into
OTIS-P (Runkel 1998) for statistical determination of pa-
rameter values using nonlinear least squares analysis. The
parameter estimates reported are those obtained from OTIS-
P. Damkohler numbers (DaI ? [? ? reach length ? (1 ?
As/A)]/velocity, Bahr and Rubin 1987) were calculated and
evaluated after the experiments to assess the reliability of
parameter estimates (Wagner and Harvey 1997).
Several hydrologic descriptors were calculated using out-
put from solute modeling results. These variables included
hydraulic residence times in the stream (Tstr? 1/?) and stor-
age zone (Tsto? As/[A ? ?]); hydraulic uptake length in the
stream channel (Shyd? Q/[A ? ?]), a measure of the average
distance a water molecule travels before entering the storage
zone (Mulholland et al. 1994); and the hydrologic retention
factor (Rh? Tsto/Shyd), the storage zone residence time per
unit of stream reach traveled (Morrice et al. 1997).
Metabolism—Whole-stream: Whole-stream metabolism
was measured using a two station diel oxygen mass balance
method similar to that described by Odum (1956) and mod-
ified by Marzolf et al. (1994). Dissolved oxygen (DO) con-
centrations were measured at both the upstream and down-
Hyporheic zone metabolism
stream stations of each chosen reach for 36 h at 15-min
intervals (YSI DO meters, Rio Calaveras both years, lower
reach of Gallina Creek), or 5-min intervals (Orbisphere DO
meters, upper reach of Gallina Creek). To account for ex-
change of oxygen between the stream and atmosphere, reaer-
ation coefficients were determined from the longitudinal de-
crease in steady-state concentrations of a dissolved volatile
tracer (propane) coinjected during the conservative tracer in-
jections described above (sensu Marzolf et al. 1994, with
modifications described in Young and Huryn 1998).
Upstream and downstream DO values along with Q, reach
travel time, and exchange with the atmosphere were used to
calculate net rates of oxygen change due to metabolism (net
ecosystem production) for each 5- or 15-min time interval
(Marzolf et al. 1994). Daily rate of ecosystem R was cal-
culated as the sum of the values of net ecosystem production
for all the time intervals during the night plus an estimate
of daytime R for each interval during the day derived from
linear regression of predawn and postdusk R values (Marzolf
et al. 1994). The sum of the daytime intervals plus estimated
daytime R was used to calculate gross primary production
(PG). To obtain areal metabolism rates, values were divided
by the bed surface area of the reach. Bed surface area was
calculated by taking the average value of wetted channel
width (taken every one or two meters along each reach)
multiplied by reach length.
Metabolism values from periods of rain were substituted
with values from a regression of prestorm and poststorm
intervals to determine the degree to which whole-stream me-
tabolism calculations were affected. The resulting calcula-
tions using modified values were then compared to original
Light was measured as photosynthetically active radiation
(PAR) photon flux density (?mol m?2s?1) (LI-COR quantum
sensor). Readings were taken at 5- or 15-min intervals and
values were integrated over daylight hours.
Linear regression was used to assess relationships between
hyporheic R, whole-stream R, and the percentage (arcsine
square root transformed) of whole-stream R contributed by
hyporheic R, and transient storage as both Asand As/A (proc
REG, SAS release 6.12 1996). For whole-stream R, the anal-
ysis was repeated using two sites from Mulholland et al.
(1997, incorporating corrections based on Marzolf et al.
1998) in addition to the four reaches from this study.
Benthic: Benthic metabolism was measured using light
and dark incubations of benthic sediment in traditional re-
circulating chambers (sensu Bott et al. 1978). Plastic trays
(n ? three or four per site) were filled with approximately
the top 2 cm of benthic sediment and inserted back into the
stream bed at least 1 month prior to being used in chamber
metabolism measurements. In 1996, the planar area of each
tray was 350 cm2and the circulating volume of each cham-
ber was 4 liters. Truckee River model benthic chambers (Al-
iquot) with a volume of 2 liters were used in 1997 with 85
cm2trays. Chambers were submerged in the stream during
all incubations, and dark conditions were achieved by cov-
ering the chambers with opaque reflective material. Cham-
bers were run for approximately 2 h in the dark followed by
2 h in the light during late afternoon. Winkler titrations
(1996) or Orion DO meters (1997) were used to measure
DO concentrations every 30 or 5 min, respectively. Three
chambers were run simultaneously and the resulting values
were averaged to obtain a value for the reach. A single
chamber at each site was run without benthic sediment in
1997. Neither chamber showed measurable change in DO
concentration during light or dark incubation, and therefore
water column metabolism was considered negligible.
Rates of metabolism were calculated from the slope of the
linear regression of DO concentration and time, resulting in
units of mg O2m?2h?1. Respiration was calculated from the
dark incubations and net primary production (PN) was cal-
culated from the light incubations. Gross primary production
was calculated by assuming a constant rate for R and adding
R to PN. Respiration values were scaled to 24 h and PG
values were scaled to the total number of daylight hours to
calculate PG/R ratios. To account for the fact that rates of PG
measured in late afternoon might not be representative of
rates over the entire day, levels of irradiance were taken into
account in scaling light incubation rates. We assumed pho-
tosaturation occurred at 400 ?mol quanta m?2s?1for pe-
riphyton in the four study reaches. This value was chosen to
result in a conservative estimate of daily PGand was based
on a review of multiple periphyton studies that found that
photosaturation typically occurred between 200 to 400 ?mol
quanta m?2s?1(Hill 1996). Mean PAR during all light cham-
ber incubations except the lower reach of Gallina Creek was
greater than 400 ?mol quanta m?2s?1(data not shown), and
we therefore assumed that rates obtained represented pho-
tosaturated conditions. Chamber rates were considered rep-
resentative of the period of daylight during the whole-stream
measurements in which PAR exceeded 400 ?mol quanta m?2
s?1. For the remaining daylight hours, values of PGwere
assumed to be 50% lower. During light incubations at the
upper reach of Gallina Creek in 1997, the DO probe in one
chamber did not function properly, and resultant values were
omitted from analyses. Sediment from the trays was ana-
lyzed for organic matter content by combustion and chlo-
rophyll a content using an acetone extraction (Wetzel and
Likens 1991) followed by measuring absorbance on a spec-
trophotometer (Hewlett-Packard HP8452A).
To test for significant differences among reaches, benthic
chamber data were analyzed using one-way analysis of var-
iance (ANOVA), with reach as the factor (four levels) and
R, PG, PN, organic matter content, and chlorophyll a content
as the dependent variables. These analyses were followed by
Bonferroni multiple comparisons where appropriate (Sokal
and Rohlf 1981).
Hyporheic: Hyporheic sediment R was measured as car-
bon dioxide generation and/or oxygen consumption in sed-
iment microcosms. Values for Rio Calaveras were taken
from rates of aerobic R obtained in autumn 1997 by Baker
et al. (2000, table 4, interface microcosms, n ? 4) and de-
termined for Gallina Creek sediment for this study in Sep-
tember 1997 (n ? 3). Hyporheic sediments and unfiltered
stream water from Gallina Creek were brought back to the
lab and stored at 4? C for less than 24 h before incubations
were performed. Approximately 0.4 L of wet sediment was
put into a 0.7 liter Plexiglas cylinder, and stream water was
Fellows et al.
Table 1. Measured physical characteristics of study reaches and parameters from solute transport modeling of conservative tracer
injections. Discharge, width, and depth are reach averages.
Reach length (m)
Discharge (Q, L s?1)
Velocity (m s?1)
Reach travel time
area (A, m2)
Storage zone cross-
sectional area (As, m2)
Storage zone exchange
coefficient (?, min?1)
time in the stream
time in the storage zone
Water uptake length
factor (Rh, s m?1)
90 21 41 25
18119 27 41
223 56 41137
Table 2. Water temperature, photon flux density of photosynthetically active radiation (PAR), and whole-stream metabolism values based
on 24-h measures of open-system oxygen balance.
(mol quanta m?2d?1)
R (g O2m?2d?1)
15.2 11.1 13.712.6
added to fill the remaining volume. Water was circulated
through the microcosms using a peristaltic pump. Water
samples were taken and analyzed for dissolved carbon di-
oxide concentration every 30–60 min, and the slope of the
regression of concentration versus time was taken as the rate
of R. A respiratory coefficient of 1.0 was used to express
rates of carbon dioxide production as DO consumption. Mi-
crocosm incubations were similar for Rio Calaveras sedi-
ment, except the microcosms were filled and buried in the
field for a period of time prior to measuring R, and changes
in DO were also measured (see Baker et al. 2000 for details).
Porosity was estimated for each sediment sample (Fetter
1994), and organic matter content was quantified by com-
bustion. Rates in units of DO consumption per volume of
dry sediment from Baker et al. (2000) were multiplied by
(1 ? porosity) for conversion to units of wet sediment, and
results from both sites are reported as volumetric rates with
units of g O2m?3wet sediment d?1. Differences in volu-
metric hyporheic R rates and organic matter content were
assessed using t-tests to compare Rio Calaveras and Gallina
Calculation of hyporheic zone contribution and size—The
areal rate of hyporheic zone R (Rhyporheic) was calculated as
the difference between areal rates of whole-stream R and
benthic chamber R (similar to the method described by Nae-
geli and Uehlinger 1997). For a given reach, the mean rate
Hyporheic zone metabolism
Table 3. Benthic sediment characteristics and metabolic rates for the four study reaches. Values are means of three chambers ? SE.
PG/R values were calculated from scaling chamber rates to 24 h (see text for details). Means within a row are significantly different
(ANOVA, Bonferroni p ? 0.05) if superscripts differ.
Chlorophyll a (mg m?2)
Percentage organic matter
R (mg O2m?2h?1)
† n ? 2.
60.2 ? 6.9ab
86.7 ? 7.2ab
120.0 ? 19.0a
49.4 ? 5.1b
0.8 ? 0.2a
49.4 ? 8.2a
57.7 ? 0.43b
107.2 ? 8.5ab
0.9 ? 0.1
1.3 ? 0.2a
57.5 ? 13.5a
160.3 ? 27.3a
217.8 ? 39.1a
1.2 ? 0.1
1.2 ? 0.03a
41.0 ? 6.7a
98.0 ? 5.4ab
139.1 ? 2.6ab
1.2 ? 0.2
1.7 ? 0.4a
42.2 ? 1.7a
44.3 ? 27.4b†
84.8 ? 27.2b†
0.6 ? 0.2†
Table 4. Values relating to calculations of hyporheic zone site. AHis the cross-sectional area of the hyporheic zone, calculated by
dividing the rate of hyporheic R determined by the difference between whole-stream R and benthic R by the rate R determined by hyporheic
sediment microcosm incubations. AHis multiplied by hyporheic sediment porosity for conversion to the cross-sectional area of interstital
water for comparison with As.
Depth of hyporheic
of R from the benthic chambers was scaled to 24 h and
hyporheic zone R was calculated with units of g O2m?2d?1.
The size of the hyporheic zone was estimated by coupling
the areal rates of hyporheic R calculated by the difference
method with independently measured rates of R obtained
from sediment microcosm incubations. This calculation as-
sumes that incubating three or four samples of hyporheic
sediment per site accurately represents hyporheic R, a pro-
cess that is spatially heterogeneous (Jones et al. 1995b). We
provide these calculations as initial assessments of hyporheic
dimensions (sensu Harvey and Wagner 2000). Areal hypor-
heic R determined from whole-stream measures was con-
verted to a linear measure with units of g O2m?1d?1by
multiplying areal rates by mean wetted channel width. The
cross-sectional area of the hyporheic zone (AH, m2) was cal-
culated by dividing this linear measure of hyporheic R by
the volumetric measure from the sediment microcosm in-
cubations (g O2m?3d?1) .
In this manner, AHrepresents the cross-sectional area of
the alluvial aquifer that is functionally within the boundaries
of the lotic ecosystem. To compare AHto As, AHwas mul-
tiplied by sediment porosity to obtain a measure of the cross-
sectional area of the saturated interstices in the hyporheic
zone. To obtain a maximum estimate of hyporheic depth for
a reach, AHwas divided by the mean wetted width of the
channel. This approach provides a spatial resolution of the
hyporheic zone as a rectangular unit beneath the wetted pe-
rimeter (similar to calculations using Asdescribed in Harvey
and Wagner 2000) and is an adequate comparative estimate
for these reaches in which stream depth is a relatively small
proportion of width.
Solute injections and modeling—The lower reach of Gal-
lina Creek exhibited the most surface–subsurface exchange
with the greatest values of As, As/A, Tstor, Rh, and the lowest
value of Shyd(Table 1). In contrast, the lowest values of As,
As/A, Tstor, and Rhwere at the short reach of Rio Calaveras
and were more than an order of magnitude lower than those
at the lower reach of Gallina Creek. Values of As/A, Tstor,
and Rhfor the upper reach of Gallina Creek were low and
similar to those of the short reach of Rio Calaveras. The rate
of exchange between the surface water and the storage zone,
?, was lowest for the long reach of Rio Calaveras, with the
other three values four to nine times greater. Discharge was
greatest at the upper reach of Gallina Creek and was more
than six times greater than that of the long reach of Rio
Calaveras, which had the lowest value of Q and the highest
value of Shyd. Damkohler numbers varied from 0.6 to 4.3,
within the range of values likely to yield reliable parameter
estimates (Wagner and Harvey 1997).
Fellows et al.
es. Each point represents the net rate of change in dissolved oxygen
concentration due to metabolism (reaeration-corrected) for a (A)–
(C) 15-min or a (D) 5-min interval. The area integrated to obtain a
value for total respiration (R, mg O2m?2d?1) is indicated by the
vertical lines and extends up to 0 for each interval. The cross-
hatched area represents the area integrated to obtain a value for
gross primary production (PG). The duration and timing of precip-
itation are indicated by gray bars along the horizontal axes and
vertical dashed lines. The date of the measurements is indicated
below each plot. (A) long reach of Rio Calaveras, (B) lower reach
of Gallina Creek, (C) short reach of Rio Calaveras, (D) upper reach
of Gallina Creek.
Whole-stream metabolism data for the four study reach-
piration (R) for the four study reaches. The height of each bar rep-
resents the value of whole-stream R, the light portion represents
benthic chamber R, and the dark portion represents hyporheic R
(calculated by difference). Percent of whole-stream R contributed
by the hyporheic zone is shown above each bar in parentheses.
Contribution of the hyporheic zone to whole-stream res-
Whole-stream metabolism—All four reaches exhibited
strong daytime signals of PGas evidenced by increased val-
ues of net change of oxygen concentration during daylight
hours (Fig. 1). Nevertheless, metabolism was dominated by
respiration, with only a single 15-min interval having a pos-
itive value of net ecosystem production (Fig. 1C). All whole-
stream PG/R ratios were much less than 1 (Table 2), with a
mean value among sites of 0.15. Whole-stream R ranged
from 2.3 to 14.7 g O2m?2d?1, with the Rio Calaveras reach-
es having the two lowest values and the lower reach of Gal-
lina Creek having the highest value. Gross primary produc-
tion ranged from 0.2 to 1.7 g O2m?2d?1, with the lowest
value at the upper reach of Gallina Creek and the highest
value at the lower reach of Gallina Creek. The lowest value
of PAR corresponded to the reach with lowest PG, at 7 mol
quanta m?2d?1for the upper reach of Gallina Creek, whereas
the highest value was 41 mol quanta m?2d?1for the long
reach of Rio Calaveras (Table 2).
Several apparent increases in respiration associated with
precipitation events can be seen as dips in the plots of me-
tabolism for the lower reach of Gallina Creek and the short
reach of Rio Calaveras (Figs. 1B and 1C, respectively).
However, these events did not affect 24 h integrated metab-
olism values substantially. Metabolism values were calcu-
lated both as the shaded areas shown in Fig. 1 and with the
intervals affected by precipitation replaced with values from
a regression of preevent and postevent values. Using a re-
gression from a point 30 min before each storm and a point
60 min after, PGfor the lower reach of Gallina Creek in-
creased by 5%, and R for the short reach of Rio Calaveras
was reduced by 6%. Metabolism values generated without
using regressions across storms were used for further anal-
Benthic chamber metabolism—Benthic production varied
substantially across reaches, whereas rates of R were rela-
tively similar (Table 3). Mean chamber PG/R ratios for all
reaches were greater than the corresponding whole-stream
ratios. Reach values of PN, PG, and chlorophyll a content
were different (ANOVA, p ? 0.010, p ? 0.026, and p ?
0.016, respectively). The short reach of Rio Calaveras had
the greatest values for both PNand PG, whereas the upper
reach of Gallina Creek had the lowest values. The upper
reach of Gallina Creek also had the lowest value of benthic
sediment chlorophyll a, and this value was significantly low-
er than that of the lower reach of Gallina Creek. Neither
benthic R nor percent organic matter differed significantly
among the four reaches (ANOVA, p ? 0.58 and p ? 0.12,
Hyporheic respiration—Areal hyporheic R was greater at
both reaches of Gallina Creek than those of Rio Calaveras
and ranged from 0.9 g O2m?2d?1to 13.7 g O2m?2d?1(Fig.
2). Similarly, the proportion of whole-stream R contributed
by the hyporheic zone was much greater for both reaches of
Gallina Creek than those of Rio Calaveras and ranged from
Hyporheic zone metabolism
(C) normalized size of the transient storage zone (As/A) or (D)–(E)
cross-sectional area of the transient storage zone (As). Whole-stream
and hyporheic R are areal rates, and percentage hyporheic R is the
portion of whole-stream R contributed by the hyporheic zone. Val-
ues for the four study reaches are shown as circles, and values for
two sites from Mulholland et al. (1997) are shown as triangles.
Plots of different measures of respiration against (A)–
40% to 93% (Fig. 2). Mean organic matter content of hy-
porheic sediment was statistically similar between sites at
0.8 ? 0.2% for Rio Calaveras and 1.1 ? 0.1% for Gallina
Creek (t-test, p ? 0.28). Microcosm respiration rates were
also similar between sites at 7.0 ? 0.9 g O2m?3d?1for Rio
Calaveras sediment (mean ? SE, Baker et al. 2000) and 6.7
? 0.8 g O2m?3d?1for Gallina Creek sediment and were
not significantly different (t-test, p ? 0.87).
Cross-sectional area of the hyporheic zone, AH, ranged
from 0.12 to 0.73 m2(Table 4), representing 5 to 21 times
the area of the above-ground stream channel. AHwas much
greater at both Gallina Creek reaches than the Rio Calaveras
reaches, with values at Gallina Creek approximately 4 times
those at Rio Calaveras. When AHwas divided by the average
wetted width of each reach, estimates of hyporheic depth
ranged from 0.14 m at the short reach of Rio Calaveras to
2.04 m at the lower reach of Gallina Creek (Table 4). Po-
rosity of Rio Calaveras sediments was 37% (n ? 3) (Baker
et al. 2000) and Gallina Creek was 39% (n ? 3). Porosity-
corrected AHstill represented from 1.9 to 8.2 times the area
of the surface stream channels. In addition, porosity-cor-
rected AHwas much greater than As, with values ranging
from 5 to 39 times greater (Table 4).
Relationships between surface–subsurface exchange and
metabolism—All three measures of R generally increased as
transient storage increased (Fig. 3). Whole-stream R and ar-
eal hyporheic R both exhibited significant relationships with
As(proc REG, r2? 0.93, p ? 0.04; r2? 0.92, p ? 0.04,
respectively). These measures of R also increased as the
magnitude of As/A increased, but the relationships were not
significant (r2? 0.87, p ? 0.07; r2? 0.86, p ? 0.07, re-
spectively). The proportion of whole-stream R contributed
by the hyporheic zone (arcsine square root transformed) was
not significantly correlated with As(r2? 0.57, p ? 0.25) or
As/A (r2? 0.49, p ? 0.30). For analyses including the two
sites from Mulholland et al. (1997), the relationship between
whole-stream R and As/A was significant (r2? 0.84, p ?
0.01), but that between R and Aswas not (r2? 0.52, p ?
Contribution of the hyporheic zone—Few studies have in-
vestigated the effects of the hyporheic zone on stream eco-
system metabolism, but all to date have demonstrated that
the hyporheic zone is a major contributor to whole-stream
respiration. Grimm and Fisher (1984) found that the hypor-
heic zone contributed 40–50% of total ecosystem R in Syc-
amore Creek, a desert stream. Fuss and Smock (1996) esti-
mated the annual contribution of the hyporheic zone to be
70% in Buzzards Branch, a sand bottom, black water coastal
plain stream. Naegeli and Uehlinger (1997) found that the
proportion of hyporheic contribution was even greater, at
74–92% of ecosystem R in the Necker, a sixth-order gravel
bed river. The hyporheic zone contributed a substantial pro-
portion of whole-stream R in all four reaches of this study
(40 to 93%) with a range very similar to that delineated by
Surface–subsurface exchange and stream ecosystem res-
piration—As pointed out by Findlay (1995), there are few
data in the literature with which to examine relationships
between ecosystem functioning and hyporheic zone hydro-
dynamics. Our work supports the conclusion of Mulholland
et al. (1997) that the magnitude of whole-stream R is related
to the extent of surface–subsurface exchange. Whole-stream
R may not be the best functional variable with which to
examine the influence of surface–subsurface exchange on
ecosystem metabolism given the capacity for benthic rates
to obscure differences in hyporheic contributions. The rela-
tionship between hyporheic R and exchange may be more
robust than that with whole-stream R, but since our study is
the first to specifically measure hyporheic R and surface–
subsurface exchange, there are even fewer data with which
to make an evaluation than for whole-stream R. The rela-
tionship between whole-stream R and As/A provided in Fig.
3A links the fluvial nature of streams to their functional sta-
Fellows et al.
tus as ecosystems and will continue to be tested as more
data become available.
The relative magnitude of hydrologic exchange at the two
study sites can also be evaluated independently of solute
transport modeling results. The contribution of the hyporheic
zone to whole-stream R was greater at Gallina Creek than
Rio Calaveras, whereas the volumetric rates of hyporheic
sediment R were similar between sites. If the contribution of
the hyporheic zone is a function of the rate of a process and
the proportion of surface water routing through the hypor-
heic zone (Findlay 1995), this suggests that the proportion
of stream water routing through the hyporheic zone was
greater at Gallina Creek. Similarly, since calculations of AH
standardize for volumetric hyporheic sediment R rates, larger
values of AHmean that a greater volume of hyporheic zone
sediments were contributing to whole-stream metabolism per
unit stream length at the Gallina Creek reaches.
Estimates of hyporheic zone size using metabolic versus
solute tracer technique—Although the hyporheic zone has
been clearly indicated as a source of transient storage in
stream ecosystems, it may be that transient storage values
determined from short-term tracer injections do not ade-
quately reflect hydrologic processes that affect metabolism
over the period of days (Harvey et al. 1996). Porosity-cor-
rected values of AHwere substantially greater than corre-
sponding values of As(5–40 times). The duration of a solute
injection determines the scale at which information about
surface–subsurface exchange is obtained (Harvey et al.
1996), and our injections generally lasted a few hours. When
metabolism measurements integrate longer duration flow
paths that do not influence short duration tracer injections,
values of AHwill be much greater than As. AHtherefore can
be used in calculating the contribution of the hyporheic zone
to other processes of interest of similar timescales, such as
nutrient cycling, if the rate of the process is known for a
defined volume of hyporheic sediment. AH, alone or stan-
dardized for stream size using A, should be a valuable cross-
site variable because it takes into account differences in rates
of both benthic and volumetric hyporheic sediment R rates.
Our work provides additional evidence of the importance
of extending stream ecosystem boundaries to include the re-
gion of surface water–groundwater exchange. The hyporheic
zone contributes substantially to ecosystem functioning in
the four reaches we studied and in many other streams (Jones
and Holmes 1996; Brunke and Gonser 1997; Boulton et al.
1998; Dahm et al. 1998). Both Asand AHare indirect mea-
sures of the size of the hyporheic zone and are useful for
cross-site comparisons and in predicting the contribution of
the hyporheic zone to whole-system measures of metabolism
and other processes. Since rates of benthic R and volumetric
hyporheic sediment R were similar across reaches, hydrology
was a significant factor determining the contribution of the
hyporheic zone to whole-system processes at these sites. The
importance of surface water–groundwater interactions to
stream ecosystem functioning highlights the need for com-
bining studies of ecosystem process with measures of hy-
drology and hydrologic modeling.
BAHR, J. M., AND J. RUBIN. 1987. Direct comparison of kinetic and
local equilibrium formulations for solute transport affected by
surface reactions. Water Resour. Res. 23: 438–452.
BAKER, M. A., H. M. VALETT, AND C. N. DAHM. 2000. Organic
carbon retention and metabolism in a near-stream groundwater-
ecosystem. Ecology 81: 3133–3148.
BENCALA, K. E., AND R. A. WALTERS. 1983. Simulation of solute
transport in a mountain pool-and-riffle stream: A transient stor-
age model. Water Resour. Res. 19: 718–724.
BOTT, T. L., J. T. BROCK, C. E. CUSHING, S. V. GREGORY, D. KING,
AND R. C. PETERSEN. 1978. Comparison of methods for mea-
suring primary productivity and community respiration in
streams. Hydrobiologia 60: 3–12.
BOULTON, A. J., S. FINDLAY, P. MARMONIER, E. H. STANLEY, AND
H. M. VALETT. 1998. The functional significance of the hy-
porheic zone in streams and rivers. Annu. Rev. Ecol. Syst. 29:
BRUNKE, M., AND T. GONSER. 1997. The ecological significance of
exchange processes between rivers and groundwater. Freshw.
Biol. 37: 1–33.
DAHM, C. N., N. B. GRIMM, P. MARMONIER, H. M. VALETT, AND
P. VERVIER. 1998. Nutrient dynamics at the interface between
surface waters and groundwaters. Freshw. Biol. 40: 427–451.
D’ANGELO, D. J., J. R. WEBSTER, S. V. GREGORY, AND J. L. MEYER.
1993. Transient storage in Appalachian and Cascade mountain
streams as related to hydraulic characteristics. J. N. Am. Ben-
thol. Soc. 12: 223–235.
DUFF, J. H., AND F. J. TRISKA. 1990. Denitrification in sediments
from the hyporheic zone adjacent to a small forested stream.
Can. J. Fish. Aquat. Sci. 47: 1140–1147.
FETTER, C. W. 1994. Applied hydrogeology, 3rd edition. Macmillan.
FINDLAY, S. 1995. Importance of surface–subsurface exchange in
stream ecosystems: The hyporheic zone. Limnol. Oceanogr.
FUSS, C. L., AND L. A. SMOCK. 1996. Spatial and temporal variation
of microbial respiration rates in a blackwater stream. Freshw.
Biol. 36: 339–349.
GORDON, N. D., T. A. MCMAHON, AND B. L. FINLAYSON. 1992.
Stream hydrology: An introduction for ecologists. Wiley.
GRIMM, N. B., AND S. G. FISHER. 1984. Exchange between inter-
stitial and surface-water: Implications for stream metabolism
and nutrient cycling. Hydrobiologia 111: 219–228.
HARVEY, J. W., AND B. J. WAGNER. 2000. Quantifying hydrologic
interactions between streams and their subsurface hyporheic
zones, p. 3–44. In J. B. Jones and P. J. Mulholland [eds.],
Streams and ground waters. Academic.
,, AND K. E. BENCALA. 1996. Evaluating the reli-
ability of the stream tracer approach to characterize stream–
subsurface water exchange. Water Resour. Res. 32: 2441–2451.
HENDRICKS, S. P., AND D. S. WHITE. 1991. Physicochemical patterns
within a hyporheic zone of a northern Michigan river, with
comments on surface water patterns. Can. J. Fish. Aquat. Sci.
HILL, W. R. 1996. Effects of light, p. 121–148. In R. J. Stevenson,
M. L. Bothwell, and R. L. Lowe [eds.], Algal ecology: Fresh-
water benthic ecosystems. Academic.
HOLMES, R. M., J. B. JONES, S. G. FISHER, AND N. B. GRIMM. 1996.
Denitrification in a nitrogen-limited stream ecosystem. Biogeo-
chemistry 33: 125–146.
JONES, J. B., S. G. FISHER, AND N. B. GRIMM. 1995a. Nitrification
in the hyporheic zone of a desert stream ecosystem. J. N. Am.
Benthol. Soc. 14: 249–258.
. 1995b. Vertical hydrologic exchange
Hyporheic zone metabolism
and ecosystem metabolism in a sonoran desert stream. Ecology
, AND R. M. HOLMES. 1996. Surface–subsurface interactions
in stream ecosystems. Trends Ecol. Evol. 11: 239–242.
, AND P. J. MULHOLLAND [EDS.]. 2000. Streams and ground
MARZOLF, E. R., P. J. MULHOLLAND, AND A. D. STEINMAN. 1994.
Improvements to the diurnal upstream-downstream dissolved-
oxygen change technique for determining whole-stream metab-
olism in small streams. Can. J. Fish. Aquat. Sci. 51: 1591–
. 1998. Reply: improvements to the
diurnal upstream-downstream dissolved oxygen change tech-
nique for determining whole-stream metabolism in small
streams. Can. J. Fish. Aquat. Sci. 55: 1786–1787.
MINSHALL, G. W., S. A. THOMAS, J. D. NEWBOLD, M. T. MONAGH-
AN, AND C. E. CUSHING. 2000. Physical factors influencing fine
organic particle transport and deposition in streams. J. N. Am.
Benthol. Soc. 19: 1–16.
MORRICE, J. A., H. M. VALETT, C. N. DAHM, AND M. E. CAMPANA.
1997. Alluvial characteristics, groundwater–surface water ex-
change, and hydrological retention in headwater streams. Hy-
drol. Process. 11: 253–267.
MULHOLLAND, P. J., E. R. MARZOLF, J. R. WEBSTER, D. R. HART,
AND S. P. HENDRICKS. 1997. Evidence that hyporheic zones
increase heterotrophic metabolism and phosphorus uptake in
forest streams. Limnol. Oceanogr. 42: 1012–1023.
, A. D. STEINMAN, E. R. MARZOLF, D. R. HART, AND D. L.
DEANGELIS. 1994. Effect of periphyton biomass on hydraulic
characteristics and nutrient cycling in streams. Oecologia 98:
NAEGELI, M. W., AND U. UEHLINGER. 1997. Contribution of the
hyporheic zone to ecosystem metabolism in a prealpine gravel-
bed river. J. N. Am. Benthol. Soc. 16: 794–804.
ODUM, H. T. 1956. Primary production in flowing waters. Limnol.
Oceanogr. 1: 103–117.
ORGHIDAN, T. 1959. Ein neuer Lebensraum des unterirdischen Was-
sers, der hyporheische Biotop. Arch. Hydrobiol. 55: 392–414.
PUSCH, M., AND J. SCHWOERBEL. 1994. Community respiration in
hyporheic sediments of a mountain stream (Steina, Black-For-
est). Arch. Hydrobiol. 130: 35–52.
RUNKEL, R. L. 1998. One-dimensional transport with inflow and
storage (OTIS): A solute transport model for streams and riv-
ers. USGS Water Resources Investigations Report 98-4018.
SAS RELEASE 6.12. 1996. SAS Institute.
SOKAL, R. R., AND F. J. ROHLF. 1981. Biometry, 2nd edition. Free-
SPEAKER, R. W., K. MOORE, AND S. GREGORY. 1984. Analysis of
the process of retention of organic matter in stream ecosystems.
Verh. Int. Verein. Limnol. 22: 1835–1841.
STREAM SOLUTE WORKSHOP. 1990. Concepts and methods for as-
sessing solute dynamics in stream ecosystems. J. N. Am. Ben-
thol. Soc. 9: 95–119.
TRISKA, F. J., J. H. DUFF, AND R. J. AVANZINO. 1993. The role of
water exchange between a stream channel and its hyporheic
zone in nitrogen cycling at the terrestrial aquatic interface. Hy-
drobiologia 251: 167–184.
, V. C. KENNEDY, R. J. AVANZINO, G. W. ZELLWEGER, AND
K. E. BENCALA. 1989a. Retention and transport of nutrients in
a 3rd-order stream: Channel processes. Ecology 70: 1877–
and transport of nutrients in a 3rd-order stream in northwestern
California: Hyporheic processes. Ecology 70: 1893–1905.
tention-transport response to nitrate loading and storm dis-
charge in a third-order stream. J. N. Am. Benthol. Soc. 9: 229–
VALETT, H. M., J. A. MORRICE, C. N. DAHM, AND M. E. CAMPANA.
1996. Parent lithology, surface-groundwater exchange, and ni-
trate retention in headwater streams. Limnol. Oceanogr. 41:
WAGNER, B. J., AND J. W. HARVEY. 1997. Experimental design for
estimating parameters of rate-limited mass transfer: Analysis
of stream tracer studies. Water Resour. Res. 33: 1731–1741.
WETZEL, R. G., AND G. E. LIKENS. 1991. Limnological analyses,
2nd ed. Springer.
WINTER, T. C., J. W. HARVEY, O. L. FRANKE, AND W. M. ALLEY.
1998. Ground water and surface water: A single resource.
USGS Circular 1139.
YOUNG, R. G., AND A. D. HURYN. 1998. Comment: Improvements
to the diurnal upstream-downstream dissolved oxygen change
technique for determining whole-stream metabolism in small
streams. Can. J. Fish. Aquat. Sci. 55: 1784–1785.
. 1989b. Retention
. 1990. In situ re-
Received: 29 February 2000
Amended: 30 November 2000
Accepted: 10 January 2001