Project Title: Application of the Marshland Upwelling System (MUS) to Treat Domestic Wastewater in Sensitive Coastal Areas
Principal Investigator(s): Kelly A. Rusch, Ronald F. Malone

NOTE: This research is dedicated to the memory of Mr. Robert E. Watson, Jr., who lost his battle to diabetes on May 3, 2002. Rob was the Ph.D. student in charge of this project. He would have defended his dissertation in December 2002.

Accomplishments
Scheduled Tasks and Progress:
Tasks 1-2: Site selection. 100% complete.
100% complete. The MUS has been installed along the northern end of the Grand Bay NERR in a tidal Juncus marsh peninsula bordering Bayou Cumbest. Two camps owned by Donald and Ingrid Glennon provide wastewater (gray and black) in addition to that produced by a public restroom (majority black water) available for public use at the adjacent boat launch.

Task 2: MUS system design. 100% complete.
A common, gravity-fed, holding tank (55 gal) collects wastewater from one camp and the outhouse. Collected water is routed to the main holding tank adjacent to the second camp via a sump pump. The main holding tank (300 gal) collects gravity-fed wastewater from the second camp and pumped waste from the holding tank.

Wastewater is sent to the injection well from the main holding tank by a timer controlled progressing cavity pump. A volumetric flow meter measures the cumulative volume of injected waste. Injection pressures are continuously monitored and recorded via a transducer and data logger.

Task 3: Fabrication and installation of the MUS system. 100% complete.
The permitting process has been completed with the Mississippi Department of Marine Resources. System installation was completed on 05/29/01. A total of 38 observation wells surrounding a 12.5 ft deep injection well have been installed. Due to tidal influences, surface waters may not be representative of undiluted system effluent. Thus, samples drawn from the 5 ft monitoring wells are considered a conservative estimate of the true effluent. Three sets of background wells at each corresponding depth were installed in areas uninfluenced by the wastewater plume. These wells indicate native salinities to help delineate the wastewater plume.

Task 4: Determination of influent wastewater characteristics. 90% complete.
Biweekly influent samples have been collected. Variable camp and outhouse usage, with peak activity during the summer months, explain the large variability. The cumulative averages of influent sampling are presented in Stephen Richardson’s Master’s Thesis.

Task 5: Determination of hydraulic conductivity of the experimental area. 100% complete.
In-situ soil hydraulic conductivities were determined on May 16, 2001 using the piezometer or slug test procedure. Well recovery to the added slug of water was estimated based on the time required for the water level to drop approximately 2 ft. This data was presented in a previous progress report.

Task 6: Determination of baseline salinity levels in the experimental area. 100% complete.
Background salinity, pH, and dissolved oxygen levels were measured in all wells twice a week for a two-month period. These values are currently being used to help delineate wastewater plume location. The cumulative averages are presented in Stephen Richardson’s Master’s Thesis.

Task 7: Determination of dispersion patterns of the freshwater plume. 100% complete.
A long-term tracer study was initiated on 7/11/01 to provide additional information with regards to plume migration within the subsurface. Diluted Rhodamine WT dye (~5 gal) was injected into the marshland subsurface and monitored through discrete well sampling. Samples were taken every four hours for the first ten days after dye injection. A weekly to bi-weekly sampling scheme followed. The results were presented in the last progress report.

Several laboratory studies (not part of the original research plan) were performed to gain a better understanding of plume dynamics. One-dimensional column studies were performed to evaluate the appropriateness of using Rhodamine WT dye as a “true” tracer. The results clearly indicate that the dye’s two isomers complete differently. Isomer 1 behaves as a conservative tracer, while isomer 2 shows a lag. From these studies, it was possible to calculate a retardation factor, which can be used in future modeling focused on plume dynamics. One-dimensional column studies were also performed using the actual raw wastewater from the field site in an effort to calculate retardation factors for fecal bacteria. Two-dimensional flume studies were performed to investigate the movement behavior of the injected wastewater under varying background salinities. The complete results of these laboratory studies can be found in the attached Master’s Thesis by Mr. Stephen Richardson.

Task 8: System evaluation. 90% complete.
A “worst case” study was performed from 1/07/02-3/30/02 to determine the system’s ability to properly treat waste under continual loading and low temperatures. During this study, artificial wastewater supplemented that available from the site to ensure wastewater was available for every injection cycle. The concentrations of the various parameters were maintained at levels exhibited by the actual wastewater (with the exception of bacteria, which were not supplemented). The influent CBOD5 concentration was maintained at approximately 250 mg/l. The 5-foot monitoring well CBOD5 concentrations during this studied average 25.4 mg/l for vector distances of 10.97 feet and 6.4 mg/l for vectors distances of 21.36 feet, indicating the system’s capability of adequately degrading the organic matter under winter conditions. Between April 1, 2002-the end of this reporting period, the system was operated at a flowrate of 0.75 gpm with an injection frequency of 15 minutes every one hour. This study extends beyond the reporting period thus, the results will be reported in the final report.

Task 9: Sample/data collection. 90% complete.
Water samples and in-situ measurements have been collected twice per month from the influent source, monitoring wells, and an upstream bayou location (background). To date, a total of 30 sampling events have been conducted since system initiation. Injection pressures are continuously monitored and recorded every minute to indicate any reductions in the injection zone permeability. Pressures are downloaded from the data logger monthly.

Task 10: Modeling plume movement. 100% complete.
Two laboratory bench-scale experiments have been constructed and are currently underway: a one-dimensional (1-D) and two-dimensional (2-D) system. The main objective of these experiments is to gain information on the transport mechanisms of bacteria in various soil types and salinities. Again, the results of these studies are detailed in Mr. Stephen Richardson’ Master’s Thesis.

The 1-D experiment focuses on measuring the degree of bacteria retention in each media type. This system consists of a column containing the media of interest, saturated at a specified salinity. In order to simulate the intermittent nature of the MUS (i.e. the pumping and resting phases of the field system), continuous and pulse injection schemes will be compared to determine the transport extent under buoyancy and injection forces.

Delineation of the wastewater plume at the field site is performed by monitoring salinity changes with depth. However, the actual appearance of the plume is unknown. The purpose of the 2-D system is to enable visual confirmation of the plume dynamics through a given media type. A conservative tracer (i.e. Rhodamine WT) will be mixed with freshwater or wastewater and injected into a saturated media. Reservoirs located at the sides of the system enable attainment of a constant salinity within the saturated media. Sampling ports along the facing wall of the system allow for continual monitoring of tracer and salinity concentrations at varying distances from the point of injection. Travel times for bacteria from the point of injection to the surface will be determined for different media types and salinity values.

Task 12: Data analysis. 65% complete.
A valid comparison of the tested injection flowrates requires hydraulic loading rates, along with influent bacterial, solids, nutrient and organic concentrations to be normalized with respect to time. The results show that loading rates were highest during the 0.75 gpm flowrate due to increased constituent concentrations in the influent. This finding suggests that the inferior performance during the 1.5 gpm study was likely due to the increased injection velocities, not the hydraulic, solids, organic or nutrient loading rates. Furthermore, these trends indicate that the MUS may be able to effectively handle greater volumes of wastewater per unit time than has been tested to date.

Task 15: Coordination of 1-2 day field day for interested parties. 100% complete.
A field day was held for Mississippi DEQ personnel.

Difficulties Encountered
Nitrogen may be removed from wastewater through the process of denitrification in which nitrate is reduced and lost as nitrogen gas. Denitrification is an anaerobic process that is dependent on nitrification. This nitrification occurs in aerobic zones. The low levels of dissolved oxygen in the primary holding tank and marsh subsurface restrict nitrification as evident by the high NH4-N and the low NO3-N concentrations of the influent. To promote nitrification within the holding tank and subsequent denitrification in the marsh soils a linear air pump was inserted at the primary holding tank. Two six inch air stones were connected to the air pump and inserted into the tank in order to achieve a dissolved oxygen level sufficient to allow nitrification to take place. The determination of long-term nitrogen removal will require at least several years of data collection. However, data collected to date indicate that nitrification is indeed proceeding in the subsurface environment as evidence by the reduction of ammonia nitrogen levels and increasing presence of nitrite nitrogen.

Redox potential probes (not part of the original research plan) were constructed in order to determine whether the oxidation and reduction state of the sand/soil matrix is sufficient to remove NO3-N from the soil through denitrification. Redox measurements will also help determine other terminal electron acceptors utilized during organic matter degradation. Additionally, redox measurements show that nitrification can occur in the upper zones of the soil.

The ability of properly assess the impact of the Juncus on the nitrification-denitrification processes occurring at the filed site is not feasible. Recall, the shallowest monitoring well is located at a 5-foot depth below ground surface. Use of shallow wells is not possible due to potential contamination by surface water percolating downward under conditions of high surface water salinity and high tides. To address this issue, separate laboratory column studies were initiated to investigate the removal of ammonia nitrogen and nitrate nitrogen by marsh soil alone and by marsh soil plus Juncus. To date, the ammonia nitrogen studies have been concluded (four month study). The results indicate that the columns with the marsh grass exhibit lower effluent concentration of ammonia than the columns without the marsh grass (the data is currently being analyzed). The results also show that the marsh grass has impacted the redox of the marsh soil to benefit the nitrification process. Complete results of these studies will be presented in the final report and as part of a Master’s Thesis focused on nitrogen removal with the MUS.

Anticipated Success in Meeting Project Objectives in Scheduled Project Period
Although the project started 3.5 months late, we are currently on schedule. With the no-cost extension, the research team will be able to develop a better understanding of nitrogen dynamics within the MUS system.

Preliminary Results
Some preliminary data is included in above discussions. The final data analysis for the fecal coliforms in included in the Master’s Thesis attached to this report. Additionally, two manuscripts have been submitted for publication in refereed journals:

Richardson, S.D. and Rusch, K.A. “Evaluation of the Marshland Upwelling System for the Removal of Fecal Coliform Bacteria from Coastal Dwelling Domestic Wastewater”, ASCE Journal of Environmental Engineering.

Richardson, S.D., Willson, C.S. and Rusch, K.A. “Implications of Using Rhodamine WT as a Wastewater Plume Tracer in the Marshland Upwelling System”, Ground Water.

Tasks and activities for next reporting period

Tasks and work plan to accomplish tasks for next reporting period
Task 4: Determination of influent wastewater characteristics.
Bimonthly samples will continue to be drawn from the main holding tank to further develop the present database of raw wastewater characteristics.

Task 8: System evaluation.
The fecal bacteria evaluation is complete. The phosphorus and biochemical oxygen demand evaluation is complete. The phosphorus evaluation is complete. Samples will continue to be collected to finish the nitrogen evaluation.

Task 9: Sample/data collection.
Biweekly sample collection will continue throughout the remaining project timeline.

Task 11: Development of operational guidelines.
Collected data will be used to develop operational guidelines for fecal coliform bacteria (done), organic matter and nutrients. Data will be reduced to hydraulic, bacterial, organic and nutrient loading rates. These guidelines will be included in the technical manual to be prepared under Task 14.

Task 12: Data analysis.
Relationships between influent and effluent substrate concentrations will be evaluated.

Task 13: Economic analysis.
The cost of implementing the MUS under various operational formats will be determined and compared with other wastewater treatment alternatives.

Task 14: Preparation of technical manual.
A preliminary MUS design, construction, and operation manual will be generated from data collected from this project. This report will be put together in the form of a refereed journal article that summarizes all data collected from this project and a comparison to historic data.

Concerns or difficulties
While all of the technical issues that have surface throughout the course of this project have been addressed, the largest and most devastating difficulty encountered to date has been the death of Mr. Robert E. Watson, Jr. on May 3, 2002. Rob was a bright young researcher who lost his battle to diabetes. Rob was a Ph.D. student in charge of the overall project. His dissertation was to encompass the complete picture of the MUS system, including the development of hydraulic, solids and organic loading rates, the potential for complete wastewater treatment and the establishment of the initial regulatory links for future certification of this technology. His premature death has left a tremendous whole in not only the research, but also the entire research team. There were three graduate students working on this research. As stated, Rob was in charge of the actual design and operational guidelines. Rob’s data collection included solids, biochemical oxygen demand and phosphorus. These parameters continue to be analyzed for the influent sample, but were discontinued from the monitoring wells effective June 30, 2002. The data, while not wholly reflected in this document is being fully analyzed and will be part of the final report and will be part of several refereed journal articles to be publish within the next eight to ten months. The phosphorus data shows results similar to other investigators. It acts as a good conservative tracer, with no transformations taking place subsurface. The biochemical oxygen demand data have indicated (at least in the short-term) that the subsurface environmental has a tremendous capacity to treat organic matter. The true test of this system can only be seen after operating it for several years to observe long-term exhaustion of the media due to continual organic loading.

Mr. Stephen Richardson was in charge of bacterial removal. This part of the research has been concluded. His data indicate the excellent ability of the system to remove bacterial organisms. His thesis compares Moss Point data to that collected from the first MUS site located in Louisiana. While the sites had distinctly different soil characteristics, they exhibited similar treatment capabilities.

Mr. Jeremy Fontenot is in charge of nitrogen removal. Jeremy will finish data collection late this fall.

Expenditures
Expenditures are in the range anticipated for work completed to date.