Stream productivity analysis with dorm—II: Parameter estimation and sensitivity

The dissolved oxygen routing model DORM, which determines productivity and respiration of a stream biological community, requires in addition to stream geometry and stream flow, parameter values for reaeration coefficients and temperature and dissolved oxygen (DO) limitations on respiration. Illustrated are the methods used to obtain numerical values in a field study at the U.S. EPA Monticello Ecological Research Station. The calculated community productivity and respiration rates are sensitive to certain input parameters and assumptions. Analysis showed that, among the many parameters needed, hydraulic residence time and reaeration coefficients should be determined with utmost care. Other sensitivity tests included temperature effects, half-saturation values and longitudinal dispersion.     

Stream productivity analysis with dorm—I: Development of computational model

A numerical dissolved oxygen routing model (DORM) is developed to determine total stream community photosynthesis (P) and community respiration rates (R) through iterative routing of two-station diel DO measurements. The model differs from existing procedures for diel curve productivity analysis; it uses the complete stream DO transport equation, including longitudinal dispersion, dependence of respiratory rate on water temperature and dissolved oxygen and wind dependent oxygen transfer through the water surface. The model, DORM, is more accurate than the traditional graphical procedures of diel curve analysis and is simple to apply. DORM is also valuable in an analysis of the results' sensitivity to the numerous coefficient and parametric assumptions required in diel curve analysis. The model is not a predictive DO model, but is designed for the interpretation of DO measurements in a stream, channel or river.     

Diurnal changes in radionuclides uptake by phytoplankton in small scale ecosystems

A continuous release of cesium-134 and strontium-85 into a simulated river system was maintained for more than 3 weeks for the purpose of evaluating the influence of photosynthetic oxygen production on cesium and strontium uptake. The flow rates and radioisotope concentration were maintained at a constant input throughout the experiment.The flow through time in the model river system was 6 h. Samples were taken from various stations every 4 h over a 36 h time interval. Temperature, dissolved oxygen, as well as other routine measurements were made. Samples were processed for plankton count and radioactivity in both the suspended phytoplankton and the water.The uptake of ¹³⁴Cs by phytoplankton was related to diurnal photosynthetic production of oxygen and a linear relationship exists between rate of change in ¹³⁴Cs concentration and the rate of net photosynthetic oxygen production.     

Separation of physical loading from photosynthesis/respiration processes in rivers by mass balance

Diurnal variations in physical and chemical concentrations, including nutrients, are observed in river ecosystems. Understanding these cycles and separating the effects of physical loading (from point and non-point sources) and biogeochemical processes are necessary for water management and the TMDL process. A chloride mass balance approach is used to separate the relative influences of physical loading and biogeochemical processes in the Bow River through Calgary, Canada, which has a significant influence on the river water chemistry. Sampling campaigns were conducted in December 2005, when minimal photosynthesis and respiration occur, and in July 2006, when river discharge is high and some photosynthesis and respiration activity is present. Samples in each campaign were collected at point source input and output along the river reach through the city every hour for a 24-hour period, allowing for time of travel. The two wastewater treatment facilities within the city contributed the majority of physical mass loading to the river, with temporal variations in effluent discharge, chloride, and nutrient concentrations. Wastewater effluent chloride to nutrient (as well as other parameter relationships) concentrations also varied diurnally. An hourly chloride mass balance was achieved, within 0.5% (average, S.D.=4.4) for December and 7.7% (average, S.D.=4.2) for July, between estimated cumulative sum values from all inputs and measured values at the river output downgradient of the city, allowing for the investigation of other parameter conservativeness. Some slight diurnal variations associated with photosynthesis and respiration were observed even with limited productivity in the river. Nitrate mass fluxes appeared to be most strongly influenced by photosynthesis and respiration processes, with phosphate being less influenced. Ammonia mass fluxes appeared to be most strongly influenced by wastewater effluent loading. Physical loading can mask or enhance biogeochemical diurnal fluctuations, creating errors in river process interpretations. Chloride was a useful tracer in the mass balance to distinguish between and assist in separating physical loading and biogeochemical processes in the river.     

Atmospheric reaeration in natural streams

The available reaeration data on natural streams have been reanalyzed to determine the relationship between the dimensionless reaeration parameter and the Reynolds number. The analysis of such data indicates that the reaeration parameter is a function of the Reynolds number and the functional relationship between these two parameters follows essentially a straight line on a log—log plot.The reaeration phenomenon in a natural stream is analogous to the dispersion mechanism in the body of water, as both of these characteristics are related to the Reynolds number. The vertical dispersion parameter or the longitudinal dispersion parameter have been found to be linearly related to the reaeration parameter on a log—log plot. The applicability of the relationship so evolved is not dependent on the regional location of the stream and takes into account both the physical and flow characteristics of the stream. The availability of the new equation for reaeration coefficient opens a new approach to evaluate the effect of photosynthesis and respiration in a natural stream.     

Evaluating eutrophication potential from river community productivity

Measurements of dissolved oxygen, temperature and solar radiation were used to estimate net community production (P-R), respiration (R), gross production (P), and the P:R ratio for six rivers during the summer of 1974. Analysis of these data and comparison with a subjective ranking of the “eutrophy” of the rivers based on nutrient and chlorophyll data showed that productivity measurements in streams can be used to evaluate eutrophication potential. No single parameter or index proved reliable, but a discriminant analysis suggested that the use of respiration and the P:R ratio in conjunction can be useful. Also, the degree of temporal variability of community metabolism was a consistently good indicator of water quality. The analyses suggest that the eutrophication potential of an “unknown” river can be determined by comparing several weeks of productivity data for summer, low-flow conditions with the data presented here.     

Quantification of oxygen fluxes in a long gravity sewer

Quantification of the oxygen fluxes in the sewer system is at present the optimal methodology to obtain information about the influence of sewers on transformations and mass balances in the urban drainage system. However, the relative and absolute values of these fluxes are practically unknown. In this work, the oxygen fluxes were quantified experimentally in a full-scale aerobic main sewer. The sewer biofilm respiration was determined with an in situ flow cell, a method that has not been used before in the sewer. The surface reaeration was determined with a gas tracer method based on the inert, non-radio-active and non-toxic gas tracer sulphur hexafluoride. In addition, the wastewater biomass respiration rate was measured. The validity of the applied methods was verified with redundant oxygen balances over a 2-km-long section. Measurement campaigns under different hydrodynamic conditions showed that the relative contribution of the biofilm, the wastewater, the reaeration and the in- and outflow with the water, all contributed significantly. However, the absolute contributions varied extensively and depended especially on the discharge. The COD conversion in the sewer could be estimated from the aerobic activity. The aerobic total degradation in the study reach was 3%. However, when extrapolated to the entire sewer net of the catchment area with 5000 PE, the COD conversion was estimated as high as 30% of the dissolved COD during the night. This indicates that the wastewater composition at the treatment plant will be strongly affected by the sewer system.     

Variations in water quality during winter in two Yukon rivers with emphasis on dissolved oxygen concentration

Winter variations in water quality were investigated in an arctic and a sub-arctic river basin in the Yukon Territory. In both cases dissolved oxygen depressions were observed during the winter, a phenomenon which occurred at all stations where water was sampled under ice. Lowest values were measured in late winter when ice cover was most extensive and when concentrations of basic cations reached highest annual values. No direct relationship was observed between oxygen levels, ice thickness and ice type. It appears that as winter progresses the stream water consists more and more of groundwater which usually contains low oxygen levels. Because of ice cover during that period little opportunity exists for reaeration and, although river sections with open water showed slightly higher oxygen concentrations than under ice samples, such open areas were not sufficiently large for reaeration. Oxygen concentrations in groundwater thus have a profound influence on oxygen concentration in the streams, but groundwater is not the only cause as other factors such as oxygen uptake by suspended material and biota possibly contribute to the low overall values. Some of the data from this study are corroborated by similar findings in Alaska suggesting that natural oxygen depressions in late winter are a widespread phenomenon in northern environments.     

Direct measurements of reaeration rates using noble gas tracers in the River Lagan, Northern Ireland

Dissolved oxygen (DO) in river systems is often depleted by polluting substances, sometimes rendering the water unfit for habitats and human use. The natural mass transfer of oxygen from the atmosphere can help to alleviate this. The reaeration coefficient, K₂, which describes the rate of oxygen absorption, is an important parameter in water‐quality modelling. Owing to difficulties in the direct measurement of K₂, values for use in water quality models are typically derived using predictive equations, but these are notoriously inaccurate. This paper presents a field method for the direct measurement of K₂ by two alternative analysis methods and includes the results of eight tests conducted on a reach of the River Lagan, in Northern Ireland. The method is based on a long‐established protocol but replaces a now‐unacceptable radioactive tracer (⁸⁵Kr) with stable noble gases. K₂₍₂₀₎ values between 3 and 80/day are reported for the test reach.     

Technique for determining the volatilization coefficients of priority pollutants in streams

The volatilization coefficients of benzene, chloroform, methylene chloride, and toluene were measured in the laboratory over a range of mixing conditions concurrently with the measurement of the oxygen absorption coefficient. Application of the constancy of the ratios between the organic solute and oxygen coefficients to the modified tracer technique for the measurement of stream reaeration coefficients permits the determination of the volatilization coefficients of these solutes for any stream or river without direct introduction of these substances into the natural waters. Volatilization coefficients expressed as the liquid-film coefficients are presented for these solutes for a number of streams and rivers of the United States.     

Uncertainty in dissolved oxygen prediction due to variability in algal photosynthesis

The contribution of variance in river community metabolism to overall uncertainty in predictions from dissolved oxygen models is evaluated using Monte Carlo simulation. Results using a modification of the Streeter-Phelps model to include a term for mean daily net production show that uncertainty in parameters used to describe community metabolism must be considered when probability distributions of dissolved oxygen in rivers are estimated. This conclusion is shown to be valid for a range of conditions and is therefore applicable to all but heavily polluted or highly turbid streams. Use of a refined version of the oxygen balance equation that accounts for diel variation in algal productivity demonstrates that variance in dissolved oxygen in rivers is likely to be much larger than that predicted by models that use mean daily values of net production.     

Cobalt interference in the “non-steady state clean water test” for the evaluation of aerator equipment—II. Occurrence and magnitude for various natural water systems

Water soluble and insoluble oxidizing compounds are formed during the non-steady state clean water test in certain natural water systems causing interference with Winkler dissolved oxygen determination. This results in pseudo high oxygen transfer efficiency of aerator devices. The interfering precipitate consists mainly of hydrated cobaltic oxide. Evaluation of the electrochemical-oxygen equivalent measurements made the presence of hydrated cobalt dioxide in the precipitate plausible. The presence of hydrogen peroxide in the filtrate suggested the formation of peroxodi- and peroxomono-sulfuric acid during the reaeration step. The pH along with the buffering capacity of a test water are the most critical water quality parameters directly affecting the magnitude of the cobalt interference. At pH values of 6.9 and below the cobalt and peroxide interference in the non-steady state clean water tests can be eliminated. Maximum cobalt interference was found to occur at pH values of 10–11.     

Cobalt interference in the “non-steady state clean water test” for the evaluation of aerator equipment—I. Causes and mechanisms

Water soluble and insoluble oxidizing compounds are formed during the non-steady state clean water test in certain natural water systems causing interference with Winkler dissolved oxygen determination. This results in pseudo high oxygen transfer efficiency of aerator devices. The interfering precipitate consists mainly of hydrated cobaltic oxide. Evaluation of the electrochemical-oxygen equivalent measurements made the presence of hydrated cobalt dioxide in the precipitate plausible. The presence of hydrogen peroxide in the filtrate suggested the formation of peroxodi- and peroxomono-sulfuric acid during the reaeration step. The pH along with the buffering capacity of a test water are the most critical water quality parameters directly affecting the magnitude of the cobalt interference. At pH values of 6.9 and below the cobalt and peroxide interference in the non-steady state clean water tests can be eliminated. Maximum cobalt interference was found to occur at pH values of 10–11.     

Macrobenthos community structural changes off Cesenatico coast (Emilia Romagna, Northern Adriatic), a six-year monitoring programme

Soft bottom macrobenthos at a station located off Cesenatico (Emilia Romagna, Northern Adriatic Sea) was investigated seasonally for six years from July 1996 to July 2002. Species composition and abundance of the community have been studied in relation to fluctuation in the water environment parameters, sediment texture patterns and mucilage, that occurred mainly in the water column at the study site. Three major Po river flow peaks occurred in November 1996, October 2000 and May 2002; after these events the community was reduced to minimum abundance values (total density<2000 individuals m(-2)). In the period between the first two episodes the river discharge remained rather low and conditions of increased salinity, lower nutrients and chl a and good oxygen saturation were experienced. The fossorial Crustacean Ampelisca diadema became dominant in the community between the first two river flow events, reaching maximum density of 10,200 individuals m(-2) and substituting the bivalve Corbula gibba, indicator of sediment instability. Species richness increased in the same period. The role of Ampelisca as a facilitator in structuring the community is discussed. Corbula gibba never recovered to initial densities, apart from an abundance peak that occurred in the summer of 2000. Faunal composition seemed to evolve slowly towards a higher degree of structural complexity (positive trend in diversity and evenness index). In the study site near-bottom mucilage events occurred in the summers of 1997, 1998, 2002; they appeared uncorrelated with the observed changes in the community structure. Multivariate analysis of community structure (MDS, ANOSIM) illustrates that community changes in this station are driven mainly by hydrographical conditions influencing sediment texture patterns and trophic resources for the benthos.     

Concerted diurnal patterns in riverine nutrient concentrations and physical conditions

Several long-term sets of hourly nitrate concentration data were obtained through deployment of a nitrate sensor in an upper reach of the River Taw, a small moorland-fed river in the South West of the UK. Examination of the data obtained during periods of low flow and the absence of rainfall in the catchment revealed the presence of marked diurnal cycles, which were in concert and negatively correlated with diurnal cycles in water temperature. After verifying that these cycles were natural, an intensive 90-h field monitoring campaign was conducted, in which river water was sampled hourly and immediately analysed in the laboratory for molybdate-reactive phosphorus (P), nitrate, nitrite, ammonium, and pH. Coincident measurements of water temperature, river discharge and solar energy were also taken at, or close to, the site. All measurements revealed diurnal patterns and all patterns were concerted. The cycles of P, nitrate, nitrite, and discharge had two maxima and minima per 24 h, while the cycle of water temperature had one, with a maximum at 20.00 and a minimum at 08.00. The amplitudes of the cycles of P and nitrate were each about 30% of the mean values, while the amplitude of the nitrite cycle was as great as 80% of the mean value on occasions. Both biological and physical mechanisms for the cycling could operate through water temperature and/or incident radiation to account for the observed phenomenon, but there remains uncertainty of which is the more important. The observations have important implications for both the accuracy of pollution assessment in rivers and the physiological rhythms of riverine organisms.     

Moment methods for analyzing river models with application to point source phosphorus

Quantitative analysis of the fate of point source pollutants in streams can be achieved by using moment analysis if the input concentration is periodic. The moment analysis permits the discrimination among various proposed models and permits the estimation of the parameters in the selected model for describing the fate of the substance in a river.Ortho phosphorus enters the Sandusky River from the Bucyrus, Ohio, sewage treatment plant with a diurnal periodic concentration change. Moment analysis is used to discriminate between reversible adsorption in the sediments and the irreversible precipitation or microbial utilization in the sediments. The parameters for the irreversible model are calculated. These include the reaction rate constant and the dispersion in the river.     

Stream productivity analysis with dorm—III: Productivity of experimental streams

Thirty-two two-station diel dissolved oxygen surveys were made in the experimental streams of the U.S. EPA Monticello Ecological Research Station. The data were analyzed by a numerical Dissolved Oxygen Routing Model (DORM) to determine total community respiratory and photosynthetic rates. The artificial channels consisted of pools and riffles and had pH of 8.0, 6.3 and 5.4. Community photosynthetic production and respiration ranged from highs of 14.8 and 10.7gm⁻² day⁻¹ to lows of 1.5 and 2.6gm⁻² day⁻¹ between June and September 1979. P/R ratios varied from 2.1 in May to 0.6 in August. Seasonal trends were affected by pH and/or age of the channel. The channels also exhibited large diel variations in photosynthetic efficiency, where dependence of photosynthetic production upon light was nonlinear and had a strong hysteresis. This hysteresis cannot be explained by physical processes and is attributed to several biological processes.     

Oxygen budgets in macrophyte impacted streams

Measurements of daily dissolved oxygen variations were made throughout a calender year on three separate segments of a stream dominated by the growth of aquatic macrophytes. Mass balance budgets, quantifying respective sources and sinks of dissolved oxygen for the study segments, demonstrated the validity of incorporating an aquatic plant respiration pattern which varied in direct proportion to dissolved oxygen concentrations. The effects of oxygen consumption by the plant population on field estimates of reaeration, daily gross primary productivity, community respiration and maximum net primary productivity in the budgeting process are evaluated.     

Causes of low oxygen in a lowland, regulated eutrophic river in Eastern England

In the River Brett, Eastern England, over the period 1955-1998 there was a significant long-term decline in dissolved oxygen (DO), as well as increases in TON (total oxidised nitrogen) and SRP (soluble reactive phosphorus). Flow decreased from 1963 to 1998. Field studies in 1998-2000 showed increased pH and a gradient of DO beneath the filamentous alga Cladophora glomerata. DO decreased through the summer. Macrophytes accounted for 45% of community respiration at the study site, while sediment accounted for 36%. In container studies, muddy sediments had the highest maximum sediment oxygen demand (SOD), but canopies of C. glomerata and Lemna minor together increased the SOD by up to 90% over control samples. During periods of high temperature, abundant growths of C. glomerata and/or L. minor would increase the SOD of organic mud in river areas with shallow, ponded water, eventually leading to anoxic conditions and the release of nutrients from the sediment. If a river had large areas of mud, these processes could dramatically affect the river's oxygen budget, and hence its ecology.     

A model of reaeration rates driven by shear flows

The reaeration rate determines the speed that the dissolved oxygen is restored to the saturation level. The reaeration rate is determined by the surface renewal rates from the friction interfaces in the water bodies, including the water/bed interface, the shear‐flow interface and the air/water interface. The formulae of reaeration rate for the air/water interface and the water/bed interface were developed in prior studies. However, no formula of the reaeration rate driven by the shear flows was developed. In this study, a mechanic model of the reaeration rate driven by the shear flows is developed to fill in the gap. The flow velocity profile in the shear flows and the Surface Renewal Theory are employed to derive the corresponding model. The predictions of the formulae for these three types of friction interfaces are compared for the same phase velocity to investigate the reasonability of the reaeration rate model for the shear‐flow interface. The predictions of the model for the shear‐flow interface are between those for the air/water interface and for the water/bed interface. The model in this study is also verified to have reasonable agreements with the experimental data. The model developed in this study can be applied for the prediction of the low soluble gases’ transfer rate between air and water in shear flows.