SPONSOR:
U. S. Army Garrison-Hawaii
PROJECT PERIOD:
09/01/00 - 03/31/01
ABSTRACT:
The Schofield Army Barracks Wastewater Treatment Plant (SBWWTP) treats wastewater from the following facilities in central Oahu, Hawaii: U.S. Army's Schofield Barracks Military Reservation, Wheeler Army Air Field, Camp Stover Housing Area, Helemano Military Reservation, U.S. Navy's Field Station Kunia, and the private Helemano Plantation Development. The plant provides secondary treatment using the conventional activated sludge process and is designed to treat an average flow of 4.2 million gallon per day (MGD). Current average daily discharge is approximately 2.2 MGD. This effluent does not consistently meet the requirements of HAR 11-62 for total coliform bacteria, due to spikes that occur during some periods. While the requirements for HAR 11-62 specify total coliform, the requirements for effluent reuse discharge specify fecal coliform. The Army is attempting to meet the requirements of their National Pollution Discharge Elimination System (NPDES) permit by upgrading the plant to consistently meet the requirements of HAR 11-62. At present, the treated effluent from the plant is discharged to an irrigation ditch operated by the Dole Foods Company (DFC). The effluent is mixed with water drawn from the Wahiawa Reservoir and used for irrigation on DFC lands. An existing easement allowing discharge into the DFC irrigation system expires in December 2001. The water is used to irrigate diversified agricultural crops such as coffee, papaya, corn, and macadamia nuts. Prior to this, the water was used to irrigate sugar cane on the same lands.
The U.S. Army Garrison-Hawaii asked the U.S. Army Engineer Research and Development Center at Waterways Experiment Station to assist with evaluating the potential impacts of using treated effluent on the groundwater. This scope of work was designed to meet a requirement made by the State of Hawaii Department of Health, Wastewater Branch (DOH). The work was performed by the University of Hawaii and the U.S. Army Engineer Research and Development Center (ERDC) Geotechnical Laboratory.
Objectives:
The purpose and objectives of this study were to evaluate the following:
1) Assess the potential impact, if any, of using irrigation water containing effluent from the Schofield Barracks Wastewater Treatment Plant (SBWWTP), with an emphasis on evaluating the movement of fecal coliform bacteria and coliphage virus through the shallow soils.
2) Estimate the ultraviolet radiation (UV) and chlorine dosages necessary to reduce the fecal coliform bacteria and coliphage virus counts in the shallow soils from irrigating with water containing effluent from the SBWWTP.
Scope of Work:
To meet the objectives listed above, it was proposed that primarily field and laboratory studies, with some computer modeling, be performed.
Laboratory analyses to evaluate the wastewater conditions measured fecal coliform bacteria and the coliphage virus as indicators. Fecal coliform bacteria were chosen as they are a more representative indicator of wastewater treatment plant performance than total coliform. The coliphage virus was chosen because it is a commonly occurring virus in wastewater, and is not commonly found as background in the environment.
Test Plots:
The field portion of the study involved the installation of test plots either on the Dole Foods Company Lands (DFC), or in an area on Wheeler Army Airfield adjacent to the SBWWTP. The test plots were built and managed with the assistance of the Water Resource Research Center (WRRC) at the University of Hawaii (UH). The location of the test plots were in an area previously irrigated for either sugar cane or diversified agriculture on the Dole Food Company (DFC) Lands. This helped to evaluate actual conditions that might occur in the DFC lands. Representatives from the U.S. Army Garrison-Hawaii assisted with negotiating site access with DFC. Each test plot was approximately 5 x 5 feet. Plot 1 received unaltered DFC ditch water, and Plot 2 received DFC ditch water supplemented with UV radiation or chlorine.
The test plots operated for 1 year, with UV disinfection for 6 months and chlorine disinfection for 6 months. The test plots were outfitted with drip irrigation lines so that water from the DFC ditch could be applied. A valve operating on a timer system was programmed to provide the correct amount of irrigation. The amount of irrigation applied was approximately 72 inches per year, which is 50% higher than average irrigation application to the DFC lands. Experience from previous studies has shown that increased application rates are necessary to obtain samples in the lysimeters. The exact amount applied may change based on discussions with DFC representatives. This irrigation water contained about 5% effluent from the SBWWTP, with the remaining 95% coming from Wahiawa Reservoir.
The test plots were planted with Bermuda grass and cut with a hand trimmer. The cuttings were removed with a hand rake. Three pan lysimeters were installed in each test plot. Pan lysimeters were used to allow movement of coliform bacteria; ceramic cup lysimeters do not allow coliform bacteria to pass through the cups. The lysimeters were installed at depths of 12 inches, 18 inches, and 24 inches. All lysimeters were installed laterally from a central access ditch to avoid disturbing the in-situ soil profile and field compaction conditions. Plot 2 was irrigated with DFC ditch water augmented with either UV or chlorine. The UV disinfection dose was 140 mW-s/cm2 with 3 banks in series. This UV doseage is representative of dosages in sewage treatment plants for R-1 reclaimed water production.
The concentration of added chlorine was 5 mg/L, which is about the maximum amount of chlorine added to effluent discharging from a treatment plant. The pan lysimeters provided sufficient sample volume for fecal coliform and virus analyses by traditional agar plate methods. At least half of the samples were also analyzed using the latest molecular biological techniques including polymerase chain reaction (PCR), reverse transcription PCR (RT-PCR), restriction fragment length polymorphism (RFLP), and 16s rRNA genetic fingerprinting. These techniques do not require large sample volumes and are capable of determining microbial community structure and the origin of the coliforms (soil, animal, human) and other bacteria. This assisted in differentiating between coliforms in the background soil and colifroms from the sewage effluent. These techniques do not apply to viruses which can only be quantified by plating techniques.
Lysimeter samples were collected approximately twice monthly, for a total of 24 sampling rounds. All 24 sample rounds were analyzed for fecal coliform and coliphage virus counts (pending sample availability). Samples were also analyzed for nitrates, phosphorus, and total dissolved solids.
This study compares the fecal coliform and coliphage virus counts of the applied irrigation water with the coliform counts after the water has percolated through the grass and soil to be captured by the lysimeters. The results from Plot 1 were used primarily to evaluate the potential for fecal coliform bacteria or coliphage virus to be transported through the soil due to the application of effluent water on irrigated lands. The results of the Plot 2 were compared to Plot 1 to evaluate what impact UV or chlorine disinfection might have on the fecal coliform bacteria counts in the effluent stream and soil.
Laboratory Analysis:
The laboratory portion of this study was used to more accurately evaluate the UV and chlorine disinfection dosage necessary to impact the fecal coliform bacteria and coliphage virus counts in Oahu soils. The effect of UV and chlorine disinfection on the entire bacterial community structure was evaluated from the 16s rRNA genetic library data. This was performed in a laboratory under controlled conditions with the soil samples placed in laboratory columns. Approximately 20 soil samples were collected with a 2-1/2" inside diameter split-spoon sampler from the DFC lands to a depth of approximately 48 inches.
These samples were returned to a laboratory and placed in soil columns. Waters with different concentrations of chlorine were added to the top of the soil samples. Samples were collected at the bottom of the soil columns after the water percolated through the columns. A vacuum was applied to the bottom (or pressure to the top) of the soil samples to induce percolation. The water samples were then analyzed for fecal coliform count, virus count, and other parameters including 16s rRNA libraries.
The effluent water was collected at the SBWWTP from the clarifier prior to the effluent entering the chlorine tank. This water, that approximately meets R-3 standards (as defined in the 1993 Effluent Reuse Guidelines), was transported to the laboratory and stored in tanks for percolation through the soil columns. The amount added to the columns was approximately the maximum amount that can be moved through the column under a vacuum and proceeded until at least 100 pore volumes have passed.
The amount of water that percolates through the columns was measured.
A total of 10 columns were constructed with water types added as listed below:
# Characteristics Purpose
1. C1 Pure tap water (Board of Water Supply water at UH) Background
2. C2 R3 water Worst case
3. C3 R3 water dosed - chlorine concentration 0.5 mg/l Minimum chlorine
4. C4 R3 water dosed - chlorine concentration 1.0 mg/l Moderate chlorine
5. C5 R3 water sand filtered and dosed chlorine, 5 mg/l R-1 chlorine dose
6. C6 Pure tap water dosed - chlorine concentration of 5 mg/l Background
7. C7 R3 water sand filtered and dosed UV of 70 mW-s/cm2 Low UV dose
8. C8 R3 water sand filtered and dosed UV of 140 mW-s/cm2 R-1 UV dose
9. C9 R3 water sand filtered and dosed UV of 200 mW-s/cm2 High UV dose
10. C10 Pure tap water dosed UV of 140 mW-s/cm2 Background
The minimum chlorine concentration of 0.5 mg/l is the amount of chlorine necessary to meet R-2 standards (with 15 minutes detention time). The moderate chlorine concentration of 1 mg/l is sometimes used to provide greater reduction of coliform in the effluent waste stream (30 minutes detention time will be used). The high dose of chlorine (5 mg/L for 120 minutes) following sand filtration is currently required for R-1 reclaimed water. The UV dose will be applied in three banks in series following sand filtration using a minimum dosage of 70 mW-s/cm2, a moderate dosage of 140 mW-s/cm2, and a maximum dosage of 200 mW-s/cm2. Most sewage treatment plants use a dose of approximately 140 mW-s/cm2. The columns containing tap water and R-3 water will be used for background and comparison with the columns having water with added disinfection. Each column run will also include a conservative tracer (chloride or bromide) to evaluate the column dispersion coefficient and soil porosity.
The object of the laboratory columns was to investigate what UV dosages and chlorine concentration would be necessary to impact the coliform bacteria and coliphage virus counts in water percolating through the soil sample. The effects on the entire bacterial community were also determined using 16s rRNA fingerprinting. The columns using tap water and undisinfected R-3 water were used for background information.
Approximately 6 additional soil samples were collected and analyzed in the laboratory for agronomic trace elements (such as manganese, iron, aluminum), moisture content, and clay content. Information from these soil samples was used to assist with evaluating fate and transport of the bacteria and viruses in the saturated/unsaturated zones.
Modeling and Numerical Analysis:
The results of the field and laboratory portions were used as input to an unsaturated zone/saturated zone model to predict the impact of coliforms and viruses on the groundwater below the DFC lands. Since depth to groundwater is 150 to 400 feet below the DFC lands, and it is not practical to complete field measurements to these depths, modeling was used to estimate impacts to groundwater below the test plots. A groundwater model together with water balance calculations was used to estimate the infiltration travel times. The modeling portion of the study allowed the data collected in the laboratory and shallow field portions to be extrapolated to a greater depth.
A FEMWATER model was built using the Department of Defense Groundwater Modeling System (GMS). The groundwater model extended from the DFC lands (just north of Schofield Barracks) to the town of Waialua. The eastern boundary was the extent of the Dole Food Lands, and the western boundary was Kaukonahua Stream. It incorporated portions of the following aquifers: The Schofield High-Level Water Body, Northern Groundwater Dam, Waialua Basal Water Body, and Waianae Dike Impounded Water Body.
It consisted of building surfaces called triangular irregular networks (TINs). These TINs were used to build the FEMWATER groundwater model. The surfaces included land surface, elevation of the bottom of the saprolite, and the freshwater/saltwater contact. The model incorporated the unsaturated zone in areas irrigated with effluent from the SBWWTP. The coliform bacteria and coliphage virus were first modeled using a conservative tracer approach assuming that they traveled at the approximate rate of water infiltrating through the unsaturated zone. Data collected from analysis of the soil samples was incorporated into the groundwater model. The use of additional models was investigated as part of this study, and reported to DOH in the progress review meetings.
PRINCIPAL INVESTIGATOR
