Category: Research Sponsored by GWRI

Principal Investigator: Rasheed Ahmad (Georgia Institute of Technology)
Principal Investigator: A. Amirtharajah (Georgia Institute of Technology)

Sponsor: GWRI
Start Date: 1994-04-01; Completion Date: 1995-03-01;
Keywords: Water Treatment, Biofilms, Microbial Detachment, Hydrophobicity, Hydrophobic Interaction, Biological Filtration, Backwashing, Bed Expansions, Air Scour

Description:

Biological Filtration may become an essential part of drinking water treatment in the United States during the next several decades. Since biological granular activated carbon (GAC) filters accumulate both biological and non-biological particles on the filter bed, the differences in the detachment of these groups of particles during backwashing will be an important aspect of operation of these filters. Among mechanisms controlling attached biofilm reactor performance, biofilm detachment is one of the aspects of the process that is least understood and studied. This study presents the results of an analysis of the detachment of particles from biofilms in terms of a microscopic force model and the evaluation of this model with results from operation of a pilot filter. Experiments on attachment and detachment were completed in a packed glass column with GAC/sand media and an indigenous bacterial population. The operation of a biological filter was simulated by using filter influent (nonchlorinated) water from the nearby Water Treatment Plant with the addition of a carbon source. Bacterial hydrophobicity is an important factor in the study of microbial attachment and detachment from surfaces. To develop a valid microscopic force model for microbial detachment, the hydrophobic force must be included in the total force calculations. The microscopic force model for detachment includes electrostatic, van der Waals, Born repulsion and hydrophobic forces. In these calculations, hydrophobic colloids produce a deeper primary minimum as compared to hydrophilic colloids. Consequently, hydrophobic bacteria are far more strongly attached to the GAC surface as compared to hydrophilic clay particles. During backwash of a biological filter, the maximum turbidity occurs close to zero time in contrast to the maximum microbial detachment which occurs later. In biological filters, the location and degree of attachment of biological (bacteria) and non-biological (clay) particles to the GAC media surface are different. Hence, their mechanics of removal and times of detachment are different. The use of different bed expansions during backwash has a bearing on microbial detachment. A sixty percent bed expansion produced maximum bacterial removal. The optimum porosity expression for traditional filter backwash can be applied to biological filters in drinking water treatment. Water washed filters even at optimum porosity build up excessive headloss over successive filter runs. Hence, fluidization water wash alone is not an effective method to clean biological filters. Biological filters for drinking water treatment need air scour. The biological filters backwashed with a combination of air plus subfluidization water flow at collapse-pulsing followed by water wash with 20% bed expansion produced not only lower headloss but also little headloss build up over successive runs. The chemistry of the backwash water has an impact on biofilm detachment. Limited experiments suggest that chlorinated backwash is detrimental to biological filtration. The biofilm detachment model developed in terms of forces at the microscopic level, explains the detachment mechanism in biological filtration and is consistent with the experimental results.

Principal Investigator: Walter G. Knisel (University of Georgia)
Principal Investigator: Matt C. Smith (University of Georgia)
Principal Investigator: Stanley R. Wilkinson (U.S. Department of Agriculture)
Principal Investigator: Matt C. Smith (University of Georgia)

Sponsor: GWRI
Start Date: 1995-04-01; Completion Date: 1995-04-01;
Keywords: GLEAMS, leaching, management, modeling, nitrogen, poultry litter, runoff, water quality

Description:

The GLEAMS model was applied to represent 6 fertilizer and broiler litter management practices on Coastal bermudagrass with multiple cuttings for hay at Watkinsville, Georgia. Observed data from a 7 year study were used to validate GLEAMS with comparisons of runoff volumes, runoff nitrogen, N03-N leached, forage yield, and nitrogen yield. Simulated runoff volumes and runoff nitrogen losses compared well with observed values for the record period. NO3-N leached and maximum monthly weighted concentrations were compared with good representation of the timing of peaks for all treatments. The model over-estimated the mass of No3-N leached; but peak concentrations were both over- and under-estimated. Forage yield was simulated very well, but potential yield was found to be a very sensitive model input parameter. The model under estimated nitrogen yield in the forage very significantly. The optimum nitrogen concentration relations under-estimated N demand, but coefficients for lush nitrogen uptake were very sensitive interactively with potential yield. Lush nitrogen uptake could result in nitrogen deficiency and cause reduced yield which does not occur in observed data. GLEAMS represented the fertilizer and broiler litter management practices, and can be used effectively for relative comparisons. However, it should not be used for quantitative predictions. Model users are urged to se local site-specific data if at all possible to fine-tune values of sensitive parameters.

Principal Investigator: Matt C. Smith (University of Georgia)
Principal Investigator: Jackie Sellers (University of Georgia)

Sponsor: GWRI
Start Date: 1994-06-01; Completion Date: 1994-06-01;
Keywords: agriculture, nonpoint pollution, nitrate contamination, wastewater

Description:

The Lake Lanier drainage basin covers approximately 665,600 acres in North Georgia and is the primary source of domestic water for the city of Atlanta. The drainage basin is located in the heart of the poultry producing region of the state with an estimated 274 million birds produced or housed within the basin. These birds could produce as much as 274,000 tons of litter per year. The objective of this study is to quantify any differences in specific water quality parameters between a forested watershed and similar watersheds whose land use is dominated by poultry production and open lands that receive applications of poultry litter. Automated water samplers were installed to collect samples from the outlet streams of four watersheds. The samples were analyzed for ammonia, nitrate, COD, total suspended solids, and orthophosphate. During the weekly sample collection, field measurements were made of pH, conductivity, and temperature. One additional sample was collected at this time for determination of BOD and fecal coliforms and streptococci. The four streams selected for monitoring were Little John, Mud Creek, East Little River, and West Little River. These sites were monitored from August, 1993 through March, 1994. The Little John site was selected as the control site, i.e. a watershed with no concentrated animal agriculture activities. Approximately 91 percent of the Little John watershed is forested and eight percent is in pasture. The East Little River and West Little sites were the same as used in a Clean Lakes Study of Lake Lanier. Mud Creek is the only site with a potential for significant influences from urban runoff and waste water treatment discharges. There are 87, 35, 43, and 0 chicken houses in the West Little, East Little, Mud, and Little John watersheds, respectively. For many of the measured parameters (dissolved oxygen, COD, BOD, total suspended solids, fecal coliforms, fecal streptococci, and temperature) there were no significant differences between watersheds. The parameters for which significant differences existed include ammonia, nitrate, electrical conductivity, pH, and orthophosphate. The data collected during this study do not provide strong evidence that the high concentrations of poultry houses and the associated land application of litter are contributing to a degradation of the water quality in the streams that were monitored. Almost all of the significant differences were associated with Mud Creek, and these differences are likely due to the discharge of municipal waste treatment plant effluent into the stream. High levels of fecal coliform and fecal streptococci bacteria have been noted previously in the streams in the Lake Lanier watershed. This study also showed high levels of bacteria in the water. The levels were high, however, for all of the sites, including Little John which is, on the whole, undeveloped.

Principal Investigator: D. E. Brune (University of Georgia)
Principal Investigator: H. D. Gunther (University of Georgia)
Principal Investigator: David E. Radcliffe (University of Georgia)
Principal Investigator: D. J. Drommerhausen (University of Georgia)

Sponsor: GWRI
Start Date: 1994-04-01; Completion Date: 1994-04-01;
Keywords: Nitrate, Animal Waste, Dairy, Electromagnetic Conductivity.

Description:

Nine dairies in north Georgia were surveyed using a ground electromagnetic (EM) conductivity meter to determine the source of high nitrate in dairy wells in the region. Ground EM conductivity increases with water and soluble salt content and has been used to map contaminant plumes from landfills and animal waste lagoons. A consistent pattern of EM readings was found at all of the dairies. There was evidence of seepage from lagoons at four of the seven dairies. However, it appeared that a more likely source of contamination in the wells was the unpaved loafing areas near the milking barns. EM readings in these areas were typically 15-20 mS m-l, whereas background readings were near 5 mS m-1. To supplement the EM readings, a total of seven groundwater observation wells were installed, five at one dairy (MO-l) and one each at two other dairies (MO-6 and PU-l). The wells at MO-l were installed at five locations: one next to the dairy barn, three in the loafing area, and one beyond the loafing area. Nitrate contents in the three wells in the loafing area ranged between 74- 96,60-105, and 83-120 mg L-1 N03-N. Concentrations in the well next to the barn and beyond the loafing area were between 12-43 and 12-16 mg L-\ respectively. The wells at MO-6 and PU-2 were located in loafing areas and had N03-N concentrations ranging from 53-83 and 105-135 mg L-1. We conclude that the loafing areas at the dairies we surveyed are a likely source of groundwater contamination by nitrate because of the high animal density and consequent high waste deposition rate.

Principal Investigator: Valentine A. Nzengung (Georgia Institute of Technology)
Principal Investigator: Evangelos A. Voudrias (Georgia Institute of Technology)
Principal Investigator: J. M. Wampler (Georgia Institute of Technology)
Principal Investigator: Charles E. Weaver (Georgia Institute of Technology)

Sponsor: GWRI
Start Date: 1993-04-01; Completion Date: 1994-03-31;
Keywords: Organoclays, Quaternary ammonium ions, Clay minerals, Montmorillonite, Sorption, Sorption kinetics, Adsorption and exchange, Aromatic hydrocarbons, Organic wastes, Wastewater treatment

Description:

The second year of a two-year research project was used to complete a series of experiments on the sorptive properties of organoclays prepared from Wyoming montmorillonite by replacing the natural exchange ions of the clay by the quaternary ammonium ions TMA (tetramethylammonium), TMPA (trimethylphenylammonium), HDTMA (hexadecyltrimethylammonium), and BDTDA (benzyldimethyltetradecylammonium). Experiments on the kinetics of sorption of naphthalene and diuron by two of the organoclays show that equilibrium is attained quickly, in most cases within one or two hours. Of the organoclays studied, TMP A-montmorillonite has the greatest sorptive ability for naphthalene and is followed in order of decreasing sorptive ability by BDTDA-, HDTMA-, and TMA-montmorillonite. Naphthalene molecules are too large to penetrate the interlayers of TMA-montmorillonite. TMA-montmorillonite is much more effective as a sorbent for benzene than it is for naphthalene. For benzene sorption, TMP A-montmorillonite is most effective, followed in order by TMA-, BDTDA- and HDTMA-montmorillonite. Diuron molecules are too large to penetrate the interlayers of TMA- and TMPA-montmorillonite, so sorption of diuron by the organoclays is increasingly effective in the order TMA-, TMPA-, HDTMA-, and BDTDA-montmorillonite, the same as the order of increasing organic carbon content of the organoclays. The presence of methanol as a cosolvent reduces the sorption coefficients roughly in the manner predicted by the solvophobic theory of Rao et al. (1985), that is, the sorption coefficients decrease approximately in log-linear fashion as the methanol content increases, but there are irregularities in some of these relationships owing to swelling of the organoclays in the presence of the cosolvent. The swelling of these organoclays in methanolwater mixtures has been quantified by x-ray diffraction analysis of samples equilibrated with the solvent mixtures. Differences in the apparent value of the empirical constant, a, for different sorbate-sorbent pairs point to other interactions not accounted for by the solvophobic theory. Experiments with radiolabelled HDTMA-montmorillonite indicate that organoclays are stable in methanol-water mixtures and that appreciable loss of organic ions by exchange with inorganic cations is appreciable only if the concentration of inorganic cations is relatively large (~1 N).

Principal Investigator: Thomas N. Debo (College of Architecture, Georgia Institute of Technology)

Sponsor: GWRI
Start Date: 1993-04-01; Completion Date: 1994-03-31;
Keywords: Computer model, Design procedures, Infiltration, Stormwater management, Urban runoff, Water quality control

Description:
This research has focused on the development of a personal computer model which was designed to assist in the design and analysis of infiltration systems. The systems included in the model are infiltration basins, infiltration trenches, dry wells, porous asphalt pavement and vegetated swales with check dams. This report outlines the data needed and procedures that must be followed to use the model, possible applications, and a computer slide show which can be used to assist in training professionals in the use of the model. Model limitations and anticipated problems in using the model are discussed. The Infiltration Computer Model is designed for use on personal computers that run MS DOS. The model is interactive in that the user is prompted for all input data by answering questions given on the computer screen. The user then has several choices in running the different design options and obtaining output on the screen or in printed format. Menus are given to allow the user to select options for design and analysis. Following are some of the other characteristics of the model. (1)Each portion of the model has on-screen editing capabilities to correct errors in data input and allow the user to change specific data values.

Principal Investigator: Robert E. Carver (University of Georgia)

Sponsor: GWRI
Start Date: 1993-04-01; Completion Date: 1993-04-01;
Keywords: Ground water, groundwater, aquifer, artesian, storativity, coefficient of storage, storage coefficient, limestone aquifer, Floridan aquifer, Georgia, Atlantic Coastal Plain, southeast Atlantic Coastal Plain

Description:

The role of storativity in production of water from the artesian portion of the Floridan aquifer in Georgia was examined from three standpoints: Expected change in water and aquifer volume under the observed historic pressure change, storage coefficients derived from historic production and observed aquifer pressure change, and the change in aquifer volume required to supply historic water production.

Sponsor: GWRI
Start Date: 1992-04-01; Completion Date: 1992-04-01;
Keywords: Reservoir Operation, Electric Power, Water Supply, Competing Uses, River Forecasting

Description:
A major complicating factor in water resources systems management is handling unknown inputs. Stochastic optimization provides a sound mathematical framework but requires that enough data exist to develop statistical input representations. In cases where data records are insufficient (e.g., extreme events) or atypical of future input realizations, stochastic methods are inadequate. This article presents a control approach where input variables are only expected to belong in certain sets. The objective is to determine sets of admissible control actions guaranteeing that the system will remain within desirable bounds. This approach is applied to the operation of Lake Lanier, a U.S. Army Corps of Engineers multiobjective reservoir and is shown to be an effective management tool. This investigation led to the following conclusions: (1) Increases of dependable energy generation are possible, albeit at the expense of wider reservoir level fluctuations and more frequent water shortages. Thus, recreation and water supply reliability are in conflict with energy generation. (2) Good quality forecasts can minimize lake level fluctuations and, at the same time, meet higher dependable energy commitments and water supply requirements. (3) Forecasting also facilitates reservoir operation with maximum outflow and generation hour restrictions. Such constraints can be used to minimize off-peak energy generation, thus maximizing overall energy value.

Technical Report

Principal Investigator: Aris Georgakakos (GWRI/Georgia Tech)
Principal Investigator: Huaming Yao (GWRI/Georgia Tech)

Principal Investigator: Seth Rose (Georgia State University)

Sponsor: GWRI
Start Date: 1993-08-01; Completion Date: 1993-08-01;
Keywords: Environmental Tritium, Stable Oxygen Isotope Ratios,Ground Water, Surface Water, Well-Pumping

Description:

Systematic temporal isotopic (δ180 and 3H) and geochemical variation was analyzed within precipitation, shallow ground water, baseflow and deeper ground water from the southeastern Piedmont Province of Georgia. These samples, acquired predominantly during 1992, provide a valuable addition to a growing set of environmental isotopic data for this region. There was a substantial degree of isotopic variation (mostly random) observed between successive precipitation events; however,only a relatively small degree of variation observed within shallow ground water, baseflow and deeper ground water acquired from four production wells. Typically δ180 and 3H values varied by ~5% within ground water between successive 1.5 – 3.0 month sampling periods. Stable oxygen isotope ratios within deeper ground water (derived from production wells) systematically decreased by ~0.5 to 1.0 per mil during the summer months. However, accompanying trends were not observed with respect to either tritium or major ion concentrations. There is little to infer from these data that ground-water pumping rates appreciably affects the age distribution (generally 15-35 years) of ground water within these small Piedmont basins. Ground water becomes isotopically well homogenized (however, not perfectly homogenized) as the result of dispersion and the exchange of water upon clay mineral surfaces.

Principal Investigator: David Wenner (University of Georgia)

Sponsor: GWRI
Start Date: 1993-04-01; Completion Date: 1993-04-01;
Keywords: oxygen isotopes, groundwaters, confined aquifer,unconfined aquifer, phreatic aquifer

Description:

A total of 127 groundwater samples were analyzed to date for their oxygen isotopic composition from 50 counties throughout the Coastal Plain of Georgia during the period from June 24, 1992 to September 17, 1992. Samples from shallow wells (<200 ft.) have measurable north to south isotopic variations, coincident with yearly mean rainfall. In general, groundwaters from the Upper Coastal Plain are about 1‰ more negative than those from the Lower Coastal Plain. Groundwaters from varying depths also have differing δ180 values, although inland and coastal areas appear to have contrasting isotopic patterns. For example, samples from cluster wells varying from 180 feet to >1200 feet deep from coastal Glynn County show oxygen isotopic enrichment with increasing depth, probably due to sea water encroachment. In contrast, samples from wells ranging from 35 feet to 900 feet deep from Worth County in south central Georgia do not have isotopic variations that correlate with depth. At both sites, however, groundwaters taken from different hydrostratigraphic levels have small isotopic variations (0.6 ‰). The present stable isotopic data set is not sufficient to be able to quantify the amount of leakage between aquifers at different hydrostratigraphic zones within the Coastal Plain.