A reaction diffusion model of C-N-S-O species in some arctic sediments

Abstract A reaction diffusion model was used to simulate the mineralization processes in an Arctic sediment. The simulation and the actual sediment were compared in relation to profiles of O₂, NO₃ and NH₄⁺. The site of particulate organic matter (POM) degradation was the single most important factor in fitting the simulation profiles to those of the sediment. It was deduced that most POM degradation occurred close to the sediment surface. When a reasonably good simulation had been obtained, the sensitivity of the model to changes in other parameters was investigated. Increases in POM degradation in the upper sediment resulted in increases in concentration of NH₄⁺ and NO₃⁻, but further increases in POM degradation created anoxic conditions below 3 mm, resulting in decreases in NO₃⁻ concentrations. The model was relatively intensive to changes in POM degradation in the lower sediment layers; increases led to more anoxic conditions and to less NO₃⁻. Increases in the C/N ratio of the POM in the lower sediment layers had little effect; increases in C/N in the upper layers led to a decrease in NH₄⁺ and NO₃⁻. The model was sensitive to changes in the first order rate constant for nitrification, but not for denitrification. Decreases in the K _m for O₂ of the nitrifying bacteria had no effect on the profiles.     

Effect of organic loading on nitrification and denitrification in a marine sediment microcosm

Abstract The effects of organic additions on nitrification and dentrification were examined in sediment microcosms. The organic material, heat killed yeast, had a C/N ratio of 7.5 and was added to sieved, homogenized sediments. Four treatments were compared: no addition (control), 30 g dry weight (dw) m⁻² mixed throughout the 10 cm sediment column (30M), 100 g dw m⁻² mixed throughout sediments (100M), and 100 g dw m⁻² mixed into top 1 cm (100S). After the microcosms had been established for 7–11 days, depth of O₂ penetration, sediment-water fluxes and nitrification rates were measured. Nitrification rates were measured using three different techniques: N-serve and acetylene inhibition in intact cores, and nitrification potentials in slurris. Increased organic additions decreased O₂ penetration from 2.7 to 0.2 mm while increasing both O₂ consumption, from 30 to 70 mmol O₂ m⁻² d⁻¹, and NO₃⁻ flux into sediments. Nitrification rates in intact cores were similar for the two methods. Highest rates occurred in the 30M treatment, while the lowest rate was measured in the 100S treatment. Total denitrification rates (estimated from nitrification and nitrate fluxes) increased with increased organic addition, because of the high concentrations of NO₃⁻ (40 μM) in the overlaying water. The ratio of nitrification: denitrification was used as an indication of the importance of nitrification as the NO₃⁻ supply for denitrificaion. This ratio decreased from 1.55 to 0.05 iwth increase organic addition.     

Feeding ecology of the freshwater detritivore Ptychoptera paludosa (Diptera, Nematocera)

1. We examined selected aspects of the nutritional ecology of larval Ptychoptera paludosa and their role in nutrient cycling in the Breitenbach, a first-order stream in Hesse, Germany.
2. Food preference experiments demonstrated significant preference for sediments with a high organic matter content and live bacteria.
3. pH was circumneutral in all sections of the gut.
4. Enzymatic activity (β-glucosidase and amino-peptidase) in different parts of the gut was measured using 4-methylumbelliferyl-β- D -glucopyranoside (MUF-Glc) and leucine-4-methylcoumarinyl-7-amide (Leu-MCA). β-glucosidase activity was highest in the hindgut.
5. The mean larval gut passage time was between 7 and 8 h.
6. The egestion rate of last instar larvae was about 0.35 mg dry weight (DW) faeces per larva h^–1 and about 1.25 mg ash-free dry weight (AFDW) faeces per mg larval AFDW day^–1.
7. Larval faeces contained at least 4–18 times more organic matter than the average in the sediments in which they were feeding, that is, larvae fed selectively, extracting organic matter from sediments.
8. P. paludosa larvae are important in the dynamics of detritus in slow-flowing reaches of the Breitenbach. They gather organic material from the sediment to a depth of 3 cm, and release it as faeces onto the sediment surface. A total of 770 g DW faeces m^–2 yr^–1, comprising about 16% organic matter, was produced by the Ptychoptera population.     

Measurement of denitrification in estuarine sediment using membrane inlet mass spectrometry

Abstract Denitrification was measured in intact sediment cores and in homogenised slurries using membrane inlet mass spectrometry. Dissolved concentrations of O₂, N₂, N₂O and CO₂ were simultaneously monitored. Using a 0.8 mm diameter needle probe, a comparison was made of the gas profiles of intact cores obtained under different conditions, i.e. with air or argon as the headspace gas and after the addition of nitrate and/or a carbon source to the sediment surface. O₂ was detectable to a depth of 1 cm under a headspace of air and the depth at which the maxima of denitrification products occurred was 1.5–2 cm. Denitrification products (N₂O, N₂) occurred in the surface layers where O₂ was above the minimum level of detectability (> 0.25 μM): diffusion of N₂ and N₂O upwards from the anoxic zone, local anaerobic microenvironments or aerobic denitrification are alternative explanations for this observation. The addition of nitrate and/or acetate increased the concentrations of N₂, N₂O and CO₂ in the sediment core. In sediment slurries, the pH, nitrate concentration, carbon source and the depth from which the sample was taken affected the rate of denitrification. Nitrogen was the sole detectable end product. Maximum denitrification occurred at pH 7.5 and at 20 mM nitrate. Denitrification was at a maximum in those slurries prepared from sections of core at 1–2 cm depth.     

Effect of algal deposition on acetate and methane concentrations in the profundal sediment of a deep lake (Lake Constance)

Abstract In the profundal sediment ot Lake Constance (143 m depth) the temperature is constant at 4 °C. Despite the constant temperature, CH₄ concentrations changed with season between about 120 μM in winter and about 750 μM in summer, measured down to 30 cm depth. The acetate concentration profiles also varied between seasons. In summer, acetate concentration reached a maximum at about 100 μM in 2 or 4 cm depth. In winter, maximal concentrations of about 5 μM were observed over the entire depth. Input of organic material in late spring may be the reason for the seasonal change of both compounds. To simulate such a sedimentation event, intact sediment cores were covered with suspensions of Porphyridium aerugenium or Synechococcus sp. The addition of the phytoplankton material resulted in a drastic increase of acetate concentrations with a maximum at 2 cm depth, similar to in situ acetate concentrations measured in summer. The same applies for CH₄ for which increased concentrations were observed down to 6 cm depth. H₂ concentrations, on the other hand, showed no distinct increase. Treatment of intact sediment cores with ¹⁴C-labeled Synechococcus cells resulted in the formation of ¹⁴C-acetate, ¹⁴CH₄ and ¹⁴CO₂. Maximum concentrations of ¹⁴CH₄ were found at 4 cm depth, i.e. just above the depth to which ¹⁴C-acetate penetrated. The results show that phytoplankton blooms may cause a seasonal variation of acetate and CH₄ in profundal sediments of deep lakes despite the constant low temperature. They also indicate that acetate is the dominant substrate for methanogenic bacteria in the profundal sediments of Lake Constance.     

Activity of methanotrophic bacteria in Green Bay sediments

Abstract Sediment pore water samples obtained from a 19 m station in Green Bay in Lake Michigan were examined for levels of ambient dissolved methane and copper, and for the potential for in situ methane oxidation by methanotrophs found within surface sediments. The in situ methane concentration in the upper oxic sediment layer ranged from 20–150 μmol · 1⁻¹ at this station. The activity of methanotrophs and the kinetics of methane oxidation in these sediments were demonstrated by the uptake of radiolabeled methane. K_s values varied between 4.1–9.6 nmol · cm³ of sediment slurry. High V_max values (12.7–35.2 nmol · cm⁻³ · h⁻¹) suggest a large population of methanotrophs in the sediments. An average methane flux to the oxic sediments of 0.24 mol · m⁻² · year⁻¹ was calculated from the pore water methane gradients. Pore water concentrations of copper in the upper sediment layer ranged from 10–120 nmol · 1⁻¹. Based upon the copper concentration, other measured parameters, and equilibrium conditions defined by WATEQF4, an estimate for dissolved free Cu²⁺ concentration of 5–38 nmol · 1⁻¹ pore water was obtained. Several factors control the rate of methane oxidation, including oxygen, methane, and the bioavailability of free Cu²⁺.     

Occurrence of phosphorus and its potential remohilization in the littoral sediments of a productive English lake

SUMMARY. 1. Vertical profiles of pH were measured at nine shallow water (<5m) locations in Esthwaite Water. These indicate strong gradients of pH near the sediment water interface suggesting a marked buffering capacity of the sediments.
2. Thirteen littoral sediment cores were horizontally sectioned and sequentially extracted (0.5 M NaHCO₃, 0.1 M NaOH, 1 M HCI) and analysed for soluble reactive phosphorus. The core sections were also analysed for total phosphorus and per cent organic content to determine the vertical and areal variability of phosphorus within the littoral sediments of Esthwaite Water.
3. The rate of release of phosphorus from intact sediment cores was measured in the laboratory as a function of the pH of overlying water, yielding the relationship log K=0.54 pH−3.94, K=mg Pm⁻²day⁻¹. The maximum release rate measured was 75 mg P m⁻² day⁻² at pH = 10.5.
4. Experiments on sediment slurries indicate that the release of phosphorus at pH 10 is rapid with approximately 50% of the total NaHCO₃+ NaOH extractable phosphorus being released within 3 h.
5. Phosphorus release from the littoral sediments may equal or exceed external sources plus hypolimnetic inputs during periods of high pH associated with times of maximum algal biomass.     

Rates of sulfate reduction and thiosulfate consumption in a marine microbial mat

Abstract The sulfur cycle in a microbial mat was studied by determining viable counts of sulfate-reducing bacteria, chemolithoautotrophic sulfur bacteria and anoxygenic phototrophic bacteria. All three functional groups of sulfur bacteria revealed a maximum population density in the uppermost 5 mm of the mat: 1.1 × 10⁸ cells of sulfate reducers cm⁻³ sediment, 2.0 × 10⁹ cells of chemolithoautotrophs cm⁻³ sediment, and 4.0 × 10⁷ cells of anoxygenic phototrophs cm⁻³ sediment. Bacterial dynamics were studied by sulfate reduction rate measurements, both under anoxic conditions (dark incubation) and oxic conditions (incubation in the light), and determination of the vertical distribution of the potential rate of thiosulfate consumption under oxic conditions. Sulfate reduction rates in the top 5 mm of the sediment were 566 nmol cm⁻³ d⁻¹ in the absence of oxygen, and 123 nmol cm⁻³ d⁻¹ in the presence of oxygen. In the latter case, the maximum rate was found in the 5–10-mm depth horizon (361 nmol cm⁻³ d⁻¹). Biological consumption of amended thiosulfate was rapid and decreased with depth, while in the presence of molybdate, thiosulfate consumption decreased to 10–30% of the original rate.     

Use of perfused cores for evaluating extracellular enzyme activity in stream-bed sediments

Abstract β-Glucosidase activity was investigated in stream-bed sediments using 4-methylumbelliferyl-β- d -glucopyranoside (MUF-β-Glc) as a model substrate. In a perfused core technique, water containing MUF-β-Glc was perfused up through sediment cores. β-glucosidase activity quantified from the release of fluorescent MUF in water discharge from the cores. At low rates of perfusion, maximum β-glucosidase activity ( V _max) in perfused sediments was similar to that in suspended (unperfused) sediments. Substrate affinity( K m)was higher in the suspended sediments. V _maxand K _m both increased when the perfusion rate was raised, although naturally-low substrate concentrations could mean that variability in perfusion rates has little effect on enzyme activity in the field. V _max was uninfluenced by whether ground or stream water was perfused through the sediments, but K _m was higher in cores perfused with groundwater. Increasing concentrations of glucose in the perfusion water resulted in a progressive inhibition of β-glucosidase activity. Although natural concentrations of glucose were low, the high turnover of enzymatically-released glucose probably means that β-glucosidase activity could be regulated by product concentration.     

Mineralization budgets in sediment microcosms: Effect of the infauna and anoxic conditions

Abstract A number of sediment incubations were set up to reproduce some of the conditions used by Kristensen and Blackburn [1] and to make a comparison with their results. There were three types of microcosm: aerobic (OX), anaerobic (AN) and aerobic with Nephtys (NOX). In addition to other measurements, dissolved organic nitrogen (DON) pools and fluxes, were measured. The sediment in this experiment contained more particulate organic matter (POM). Nephtys (NOX) had the same effect as Nereis in increasing the rate of mineralization of POC and PON, compared with the OX-cores (2.1 and 2.6 times, respectively). Again, the AN-cores had a higher mineralization rate (loss of POM) than that of the OX-cores, but in addition, mineralization in NOX-cores was not significantly different from AN-cores. It was thus confirmed that anoxic mineralization could be as high, or higher, than the oxic process. Both the temporal patterns of O₂-and and CO₂-fluxes and their magnitudes were very similar to those reported earlier. This contrasts with the higher loss of POM in the present experiment. However, the loss of C in DOC (associated with the measured DON) can account for the extra POM loss. The pore-water profiles of σCO₂ and NH₄⁺ were similar to those in the earlier report, and the fluxes of σCO₂, O₂, NH₄⁺ and NO₃⁻ followed the same temporal pattern.     

Microbial activity associated with seston in headwater streams: effects of nitrogen, phosphorus and temperature

SUMMARY 1. The influences of temperature and dissolved nitrates and phosphates on microbial activity associated with suspended fine particulate organic matter (seston) were evaluated in four headwater streams in the southern Appalachian Mountains.
2. Temperature manipulations of ± 5°C always induced significant changes in [¹⁴C] glucose mineralization (ANOVA; P <0.05) and [³H]thymidine incorporation (ANOVA; P <0.05).
3. Nutrient amendments of 1.0 mg NO₃ I⁻¹ and 0.05 mg PO₄I⁻¹ induced no significant alterations in bacterial mineralization of [¹⁴C]glucose (ANOVA; P >0.05) or incorporation of [³H]thymidine (ANOVA; P >0.05) in short-term (i.e. 3 h) experiments.
4. Microorganisms attached to refractory particulate organic matter do not appear to be limited by nitrogen or phosphorus even in streams with ambient nutrient concentrations as low as 0.06 mg NO₃ I⁻¹ and <0.03 mg PO₄ I⁻¹.
5. Our results indicate that variations in water temperature resulting from diurnal and seasonal temperature fluctuations, forest clear-cutting, and catchment elevation and aspect can have marked effects upon microbial activity and production, while short-term alterations in nutrient regime appear to have no significant effect on microbial activity associated with seston.     

Sulphate reduction in oxic and sub-oxic North-East Atlantic sediments

Abstract Oxic and sub-oxic N.-E. Atlantic sediments were examined for sulphate-reducing activity. Oxygen and/or nitrate reduction are probably the dominant mineralisation processes in the abyssal plain sediment studied. A low rate of sulphate reduction (0.1 nmol SO²⁻₄/ml/day) was recorded in the surface 5 cm of the continental slope sediment, together with the presence of a range of sulphate-reducing bacteria (SRB). A higher activity of sulphate reduction (2.2 nmol SO²⁻₄/ml/day) occurred in the continental shelf sediment which led to a small decrease in pore water sulphate and an increase in titration alkalinity. This sediment contained approx. 10²–10³ acetate, lactate and propionate oxidising SRB/ml. No low- M _r organic acids were detected in these sediments. However, amendment with 75 μM acetate stimulated sulphate-reducing activity in the shelf sediment.     

Effects of soil frost on nitrogen net mineralization, soil solution chemistry and seepage losses in a temperate forest soil

Freezing and thawing may alter element turnover and solute fluxes in soils by changing physical and biological soil properties. We simulated soil frost in replicated snow removal plots in a mountainous Norway spruce stand in the Fichtelgebirge area, Germany, and investigated N net mineralization, solute concentrations and fluxes of dissolved organic carbon (DOC) and of mineral ions (NH₄⁺, NO₃⁻, Na⁺, K⁺, Ca²⁺, Mg²⁺). At the snow removal plots the minimum soil temperature was −5 °C at 5 cm depth, while the control plots were covered by snow and experienced no soil frost. The soil frost lasted for about 3 months and penetrated the soil to about 15 cm depth. In the 3 months after thawing, the in situ N net mineralization in the forest floor and upper mineral soil was not affected by soil frost. In late summer, NO₃⁻ concentrations increased in forest floor percolates and soil solutions at 20 cm soil depth in the snow removal plots relative to the control. The increase lasted for about 2–4 months at a time of low seepage water fluxes. Soil frost did not affect DOC concentrations and radiocarbon signatures of DOC. No specific frost effect was observed for K⁺, Ca²⁺ and Mg²⁺ in soil solutions, however, the Na⁺ concentrations in the upper mineral soil increased. In the 12 months following snowmelt, the solute fluxes of N, DOC, and mineral ions were not influenced by the previous soil frost at any depth. Our experiment did not support the hypothesis that moderate soil frost triggers solute losses of N, DOC, and mineral ions from temperate forest soils.     

Seasonal changes in vertical distribution, biomass and faecal production of tubificids in the profundal region of a shallow Japanese lake

Two tubificid species Limnodrilus hoffmeisteri and L. claparedeianus formed more than 93% of the total number of oligochaetes in the profundal. Limnodrilus spp. worms were found down to 33 cm in the sediment but in great numbers in the upper zone in June and October. Worms confined to the top 15 cm of sediment accounted for 53-92% of the total number. There were two annual maxima in population density and biomass, one in late spring (66000 inds m⁻², 17 g wet wt m⁻²) and the other in mid autumn (97000 inds m⁻², 176 g wet wt m⁻²). Two regression lines describing the effect of temperature on faecal production rate were obtained; Log F = 0.0604 T (°C) −0.7660 (below 15°C), Log F = 0.0266 T – 0.2170 (above 15°C). In total 26.8 kg dry wt m⁻² of sediment was defecated annually by Limnodrilus spp. The sediment in the 0–10 cm stratum may pass through the guts of the worms 2.3 times a year. Sedimentation rates in profundal region were very low with respect to the faecal production rates of the tubificids.     

Organic matter mineralization in an organic-rich sediment: Experimental stimulation of sulfate reduction by fish food pellets

Abstract The combined effects of organic matter additions and temperature on short chain fatty acid (SCFA) turnover, sulfate reduction and nutrient accumulation were examined in an organic-rich fish farm sediment. Fish food pellets, which contribute significantly to the organic matter loss from fish farms, were added to surface sediment at three loadings (2.8; 14.0; 28.0 mg ww g⁻¹ ww sediment; equivalent to organic matter loadings measured during fish farming) and incubated for 30 days in anaerobic bags at 5°C and 15°C. SCFA accumulated to high levels (acetate up to 85 mM, propionate up to 17 mM, butyrate up to 25 mM) in sediments amended with food pellets, and sulfate reduction was stimulated up to 30 times relative to unamended sediments. Sulfate reducers appeared saturated with substrates (SCFA) even in the lowest additions. A low C/N ratio (0.4–1.8) of the major mineralization products (TCO₂ and NH₄⁺) indicated preferential nitrogen mineralization in amended sediment compared with the total particulate pool (C/N = 8.8–11.9) and added food pellets (C/N = 8.4).     

The effect of current velocity and sediment on the drift of the mayfly Ephemerella subvaria Mcdunnough

SUMMARY. Experiments conducted in an artificial stream showed that significantly more nymphs drifted from an inorganic substrate at a mean current velocity of 28.5 cm s⁻¹ than at 18.5 cm s⁻¹. Drift density, however, was not affected. Disproportionately large numbers of nymphs drifted while current velocities were being increased from 18.5 to 28.5 cm s⁻¹.
Both drift numbers and drift density were greater in turbid water, after the addition of large amounts of inorganic sediment, than under clear-flowing conditions during dark periods but not in the light. The interaction of increasing current velocity and sediment levels resulted in a significantly greater number of drifting nymphs under lighted conditions.
Minor spates which do not seriously disturb the stream bed may initiate significant increases in macroinvertebrate drift.     

In situ stimulation of bacterial sulfate reduction in sulfate-limited freshwater lake sediments

Abstract By adding sulfate in the form of solid gypsum, it was possible to transform in situ a predominantly methanogenic sediment ecosystem into a sulfate-reducing one. The concentrations of sulfate, sulfide, methane, acetate, propionate, soluble iron, and manganese were determined in the porewater before and after the transition. Although sulfate was no longer limiting, acetate and propionate continued to accumulate and reached much higher concentrations than under sulfate-limited conditions. Metabolic activities of fermenting bacteria and of sulfate reducers, which belong to the group that incompletely oxidizes organic material, might be responsible for the increased production of volatile fatty acids. The elevated concentrations of soluble Fe(II)²⁺ and Mn(II)²⁺ observed in the porewater stem from iron and manganese compounds which may be reduced chemically by hydrogen sulfide and other microbially produced reducing agents or directly through increased activities of the iron and manganese reducing bacteria. In the horizon with high sulfate-reducing activities the methane concentrations in the porewater were lower than in non-stimulated sediment regions. The shape of the concentration depth profile indicates methane consumption through sulfate reducing processes. The in situ experiment demonstrates the response of a natural microbial ecosystem to fluctuations in the environmental conditions.     

An application of the plaster dissolution method for quantifying water velocity in the shallow hyporheic zone of an Appalachian stream system

1. A method for quantifying interstitial water velocity based on the dissolution rate of plaster of Paris standards was developed as part of a study of vertical, longitudinal (1–4 order sites) and seasonal variation in the biotic and physical characteristics of the shallow hyporheic zone (0–30 cm) of a headwater stream system in West Virginia, U.S.A.
2. A calibration model was developed using a water velocity simulation tank to relate mass loss of plaster standards to water velocity and temperature. The model was then used to calculate water velocity through artificial substrata embedded in the shallow hyporheic zone of four stream reaches based on in situ mass loss of plaster standards.
3. Water velocity in the hyporheic zone increased with stream order, was highest in early spring and winter during high stream base flows, and decreased with depth into the substratum. There was a strong interaction between depth and season: during periods of high stream discharge, water velocity through the upper level of the shallow hyporheic zone (0–10 cm into the substrate) increased disproportionately more than velocity at greater depths. Mean interstitial velocity in March ranged from 0 cm s^–1 in the lowest level (20–30 cm) to 3.5 cm s^–1 at the upper level (0–10 cm) at the first‐order site, and from 2.5 cm s^–1 (20–30 cm) to 9.5 cm s^–1 (0–10 cm s^–1) at the fourth‐order site. Gradients in stream discharge and sediment permeability accounted for treatment effects.
4. Use of calibrated data improved the ability to resolve among‐season differences in interstitial water movement over the use of uncalibrated mass loss data. For some applications of the plaster standard method, empirical calibration may not be necessary.     

Spatial and temporal variation of microbial respiration rates in a blackwater stream

1. The extent of spatial and temporal variation of microbial respiration was determined in a first-order, sand-bottomed, blackwater stream on the coastal plain of south-eastern Virginia, U.S.A.
2. Annual mean respiration rates (as g O₂ m^–3 h^–1) differed significantly among substrata: leaf litter, 12.9; woody debris, 2.4; surface sediment, 0.8; hyporheic sediment, 0.4; water column, 0.003. Rates associated with wood were higher than those with leaves when expressed per unit surface area.
3. Highest respiration rates on leaves, wood and in the water column occurred during the summer, whereas rates in the sediments were greatest during the late autumn and winter. Water temperature, as well as particulate organic matter and nitrogen content of the substrata, was correlated positively with respiration rates.
4. A stepwise multiple regression showed that temperature and nitrogen content together explained 88% of the variation in respiration rates of leaves and wood. In contrast, particulate organic matter content and nitrogen content explained 89–90% of the variation in respiration in the sediments. Although water temperature was a significant factor in the sediment multiple regressions, its addition as an independent variable improved the regression models only slightly.
5. Annual mean respiration in the stream channel, based on the proportional amount of respiration occurring associated with each type of substratum during each month, was 1.1 kg O₂ m^–2 yr^–1. Seventy per cent of respiration in the stream occurred in the hyporheic zone, 8–13% occurred in the surface sediment, leaf litter or woody debris, and < 1% occurred in the water column. Approximately 16% of total detritus, or 40% of non-woody detritus, stored in the stream during the year was lost to microbial respiration.     

Spatial and temporal variation of microbial respiration rates in a blackwater stream

1. The extent of spatial and temporal variation of microbial respiration was determined in a first-order, sand-bottomed, blackwater stream on the coastal plain of south-eastern Virginia, U.S.A.
2. Annual mean respiration rates (as g O₂ m^–3 h^–1) differed significantly among substrata: leaf litter, 12.9; woody debris, 2.4; surface sediment, 0.8; hyporheic sediment, 0.4; water column, 0.003. Rates associated with wood were higher than those with leaves when expressed per unit surface area.
3. Highest respiration rates on leaves, wood and in the water column occurred during the summer, whereas rates in the sediments were greatest during the late autumn and winter. Water temperature, as well as particulate organic matter and nitrogen content of the substrata, was correlated positively with respiration rates.
4. A stepwise multiple regression showed that temperature and nitrogen content together explained 88% of the variation in respiration rates of leaves and wood. In contrast, particulate organic matter content and nitrogen content explained 89–90% of the variation in respiration in the sediments. Although water temperature was a significant factor in the sediment multiple regressions, its addition as an independent variable improved the regression models only slightly.
5. Annual mean respiration in the stream channel, based on the proportional amount of respiration occurring associated with each type of substratum during each month, was 1.1 kg O₂ m^–2 yr^–1. Seventy per cent of respiration in the stream occurred in the hyporheic zone, 8–13% occurred in the surface sediment, leaf litter or woody debris, and < 1% occurred in the water column. Approximately 16% of total detritus, or 40% of non-woody detritus, stored in the stream during the year was lost to microbial respiration.