Report of Findings from the March 7 13, 2014 Study Period. FDA Technical Assistance and Training Project

Transcription

1 Evaluating the Dilution of Wastewater Treatment Plant Effluent, Treatment Efficiency, and Potential Microbial Impacts on Shellfish Growing Areas in Bayou La Batre, AL Report of Findings from the March 7 13, 2014 Study Period FDA Technical Assistance and Training Project Reported by: U.S. Food and Drug Administration Center for Food Safety and Applied Nutrition Office of Food Safety Shellfish and Aquaculture Policy Branch Field Engineering and Data Analysis Team College Park, MD

2 September 2015 TABLE OF CONTENTS 1.0 INTRODUCTION Executive Summary Study Objectives StudyArea Background FDA Guidance on Establishing Closure Zones for WWTP Discharges Description of Bayou La Batre WWTP General Description of Study Design METHODS... See Appendix RESULTS Drogue Study and Preliminary Dye Study Weather Conditions Dye Injection Travel Time Dye Readings at Stations Dye Readings by Tracking Fluorometers Profiles of Dye at Depth Projections for Different Wastewater Treatment Plant Flows Microbiological Analysis of WWTP Influent and Effluent Short Term Failure Dilution and Anticipated Fecal Coliform Concentrations Determination of 400:1 and 1000:1 and Higher Dilutions (WWTP Failure) CONCLUSIONS AND RECOMMENDATIONS.14 2

3 APPENDIX 1 METHODS 2.1 Dye Standard Preparation and Fluorometer Calibration 2.2 Drogue Study and Preliminary Dye Study 2.3 Dye Injection 2.4 Dye Tracing 2.5 Dilution Analysis Dye Readings from Submersible Fluorometers 2.6 Microbiological Analysis of Wastewater APPENDIX 2 - LIST OF FIGURES AND TABLES Figure 1: Map of Station Locations, Diffuser, and Classified Growing Areas Figure 2: Velocity and Travel Time Estimates Based on Dye Data Figure 3: Station 1 WET Labs Data Figure 4: Station 2 WET Labs Data Figure 5: Station 4 WET Labs Data Figure 6: Station 5 WET Labs Data Figure 7: Station 6 WET Labs Data Figure 8: Dilution Zones and Accumulated Dye Tracking Results for a 0.75 MGD Flow Based on March 10 13, 2014 Study Period Figure 9: Dilution Zones and Accumulated Dye Tracking Results Projected for a 1.0 MGD Flow Based on March 10 13, 2014 Study Period Figure 10: Dilution Zones and Accumulated Dye Tracking Results Projected for a 2.0 MGD Flow Based on March 10 13, 2014 Study Period Figure 11: Dilution Zones and Accumulated Dye Tracking Results Projected for a 3.0 MGD Flow Based on March 10 13, 2014 Study Period Table 1: Influent from the Bayou La Batre WWTP FC and MSC Levels Table 2: Pre-Disinfected Effluent from the Bayou La Batre WWTP FC and MSC Levels Table 3: Post-Disinfected Effluent from the Bayou La Batre WWTP FC and MSC Levels Table 4: Influent and Effluent Samples from the Bayou La Batre WWTP Human Virus Levels APPENDIX 3 UPDATED AND ADDITIONAL FIGURES Figure 1: Dye Tracking Results for a 0.75 MGD Flow Based on March 10 13, 2014 Study Period Figure 2: Dye Tracking Results Projected for a 1.0 MGD Flow Based on March 10 13, 2014 Study Period Figure 3: Dye Tracking Results Projected for a 2.0 MGD Flow Based on March 10 13, 2014 Study Period Figure 4: Dye Tracking Results Projected for a 3.0 MGD Flow Based on March 10 13, 2014 Study Period Figure 5: Distance vs Dilution East of Outfall (minimum dilutions shown) Figure 6: Distance vs Dilution West of Outfall (minimum dilutions shown) Figure 7: Distance vs Dilution Regression East of Outfall Figure 8: Distance vs Dilution Regression West of Outfall Figure 9: 100 Meter Buffer Frequency of Outfall 3

4 1.0 INTRODUCTION 1.1 Executive Summary A hydrographic dye study of effluent from the Bayou La Batre (Buford Bryant) wastewater treatment plant (WWTP) was conducted on March 6-13, 2014 in Grand Bay and Portersville Bay. Staff from the U.S. Food and Drug Administration (FDA), the Alabama Department of Public Health (ADPH), and Alabama s Department of Marine Resources (DMR) participated in the study. Six fixed fluorometer stations were deployed at various locations along the anticipated path of the dye-tagged effluent to determine the dye concentrations and associated dilutions of the effluent at each location. Boat-towed tracking fluorometers were used to measure the surface level of dye-tagged effluent near each cage and in other parts of Grand Bay and Portersville Bay. Microbiological analyses of fecal coliforms (FC), male-specific coliphage (MSC), norovirus (NoV) genogroup I (GI) and genogroup II (GII), and adenovirus (AdV) were conducted for samples from the WWTP. The results of the microbiological analyses and the dye study are presented in this report. Based on boat tracking of the dye, the dye-tagged effluent moved at a velocity of 0.70 mph on March 10, 2014 and the estimate of effluent travel time from the diffuser to the aquaculture sites near Fowl River is 7.1 hours. The estimate of effluent travel time from the diffuser to the aquaculture sites in Grand Bay is 4.3 hours. If a prohibited area based on 400:1 dilution zone is employed around the diffuser at a radius of 1.49 miles, as recommended in this report, the travel time from the diffuser to the border of the prohibited area would be 2.1 hours. Wind-driven circulation was a significant factor in the velocity and movement of the dye-tagged effluent throughout the study. The average wind speed on March 10, 2014 was only 5 mph out of the southwest. The high wind speed was 17 mph. Under different wind speeds and directions, the dye tagged effluent may move even faster towards the Fowl River or Grand Bay aquaculture sites. The ability to close the conditionally approved growing area in less than 2.13 hours, or at least any parts of the growing area that are being harvested at the time of a WWTP failure - the aquaculture sites in Grand Bay in less than 4.3 hours and the aquaculture sites near Fowl River in less than 7.1 hours - may be necessary in the event of a raw sewage or UV disinfection failure at the WWTP. FDA recommends that a communication system be set up between ADPH, the Alabama Department of Conservation and Natural Resources (ADCNR), the Bayou La Batre WWTP, and the licensed harvesters of the growing area to ensure that harvesters are notified to stop harvesting in a sufficient period of time in the unlikely event of a failure at the WWTP. The results of the dye study indicated that surface dye levels were the same as bottom dye levels near the stations during the period of tracking. Since a full steady state analysis could not be done with all of the station fluorometers due to the non-continuous flows of the WWTP and the significant impact of wind-driven circulation on the movement of dye, the overall findings of this report emphasize the surface dye concentrations and corresponding dilutions detected via boat tracking. The minimum level of dilution recommended by FDA to mitigate the risk of viruses in treated effluent in a general scenario is 1000:1 dilution. Based on the dye study data, the 1000:1 dilution line at the surface occurs at a radius of 2.01 miles around the diffuser for a projected 4

5 flow of 1.0 MGD (the current average flow at the WWTP) as shown in Figure 9 of this report. In the case of the Bayou La Batre WWTP, the facility has sand filtration, UV disinfection, and a number of back-up systems. The microbiological results of samples from the WWTP show that the plant is efficient at removing FC indicator bacteria and meets its permit requirements. FDA testing of effluent from the WWTP did not indicate a concern for viruses or viral indicators. Therefore, if the quality of effluent does not degrade in the future as flows increase and the WWTP becomes older, a 400:1 dilution zone can be applied around the diffuser, since this is the level of dilution FDA recommends for plants with UV disinfection that demonstrate exceptional treatment capabilities. The 400:1 dilution line at the surface occurs at a radius of 1.43 miles around the diffuser for a projected flow of 1.0 MGD. 1.2 Study Objectives The study objectives were to: (1) assess the dilution, time of travel, and dispersion of effluent in Grand Bay and Portersville Bay; (2) determine the bacterial and viral loads in raw, untreated wastewater and in pre-disinfected effluent and the efficiency of the WWTP, which uses sand filtration and UV disinfection, to reduce these loads before discharge; (3) provide guidance to the Alabama Department of Public Health (ADPH) regarding the sizing of the prohibited area around the new WWTP diffuser based on dilution of effluent; and (5) further develop/expand the database that supports the level of effluent dilution needed to achieve the safety of shellfish harvested in proximity to WWTPs as part of FDA s dilution guidance, particularly for WWTPs with UV disinfection. 1.3 Study Area Background The study was performed in Area II, which is an area designated as conditionally approved that includes portions of Portersville Bay and Grand Bay. A comprehensive sanitary survey of the area was conducted in 2009 and will be performed again in 2019 with annual reevaluation reports in the interim. A prohibited zone is currently established around the old WWTP outfall in Portersville Bay and a revised prohibited zone will be established around the new outfall based on the results of this report. Bayou Coden and the portion of West Fowl River east of Channel Marker 28 are also designated as prohibited. The aquaculture sites in Area II are permitted to harvest year round. Temporary closures for conditionally approved harvest areas are based on trigger points which are representative of conditions that exist when shellstock are most probably unsafe for human consumption. The present management plan closes a shellfish harvesting area during any of the following conditions: 1) when there is discharge or spillage of any substance that is considered hazardous to public health, 2) when there is the presence of biotoxins in concentration levels deemed to be detrimental to the public health, 3) when the stage of the Mobile River reaches eight (8) feet at Barry Steam Plant, Bucks, Alabama as measured and reported by the Army Corps of Engineers, Mobile District Water Information Center, or 4) any other event such as a tropical storm, hurricane, tidal surge, etc. that could pose any significant potential public health threat. Closures do not occur based solely on rainfall levels. Closures resulting from discharges or spillage of hazardous substances and unusual weather events (tropical storms, hurricanes, heavy rains greater than 5 inches) are defined as Emergency Closures. 5

6 At the time of the study, there were no emergency closures in place and Area II was open, but the aquaculture sites to the east near Fowl River were not being harvested. ADPH reported to FDA during the study that Area II can be closed within 12 hours in the event of a failure at the Bayou La Batre WWTP. 1.4 FDA Guidance on Establishing Closure Zones for WWTP Discharges In consideration of Section II, Chapter E(5) (Prohibited Classification Wastewater Discharges) of the National Shellfish Sanitation Program Model Ordinance, which notes that the determination of the size of a prohibited zone around a WWTP outfall shall include the wastewater s dispersion and dilution, and the time of waste transport to the growing area where shellstock may be harvested (iii), FDA has provided guidance to state shellfish control authorities to size prohibited zones around WWTP outfalls according to the following scenarios: Scenario 1: In consideration of effluent discharged from a WWTP under failure conditions (such as a loss of disinfection), the prohibited zone should provide a sufficient amount of dilution to dilute the effluent discharged under failure conditions to the fecal coliform standard of 14 FC/100 ml within the boundaries of the prohibited zone. OR Scenario 2: In order to reduce the size of the prohibited zone, a conditionally approved zone may be operated if a factor of at least a 1000:1 dilution of effluent is achieved within the prohibited area to mitigate the impact of viruses, and there is a sufficient amount of time to close the conditional area to the harvesting of shellfish before the effluent discharged at the onset of a failure can travel to the boundaries of the prohibited zone Note: the additional area beyond the prohibited zone to be closed under WWTP failure conditions should provide a sufficient amount of dilution to dilute the effluent discharged under failure conditions to the fecal coliform standard of 14 MPN/100 ml within the closed (due to failure) zone (consistent with Scenario 1). Over the years, wastewater treatment technologies have improved. During this time FDA has maintained a conservative position recognizing that any WWTP may remain subject to failure. FDA recognizes that with advancements in technologies, including improved monitoring and alarm systems for a treatment bypass or loss of disinfection, it may be possible to operate a conditional area as outlined in Scenario 2 above. This allows additional shellfish growing areas to be harvested under certain conditions. When a WWTP is operating normally, disinfection has been shown to be effective in reducing the coliform bacteria group (fecal coliform and total coliform) to levels below shellfish harvesting standards as can be seen in WWTP permit records kept in accordance with the Environmental Protection Agency (EPA) National Pollutant Discharge Elimination System (NPDES) Program. However, human enteric viruses such as noroviruses and hepatitis A virus are more resistant to disinfection and thus are not reduced to the same degree as the coliform bacteria group. In an effort to mitigate the risk of contaminating shellfish with viruses, FDA has recommended a 1000:1 dilution as described in Scenario 2 as the minimum zone of dilution 6

7 needed when the WWTP is operating under normal conditions, unless an alternative approach is well supported by data. One of the alternative approaches recommended by FDA is to use 400:1 dilution for plants with UV disinfection that demonstrate exceptional treatment capabilities and the absence of significant levels of viruses and viral indicators in treated effluent. In the case of the Bayou La Batre WWTP, the use of a 400:1 dilution zone is supported by the data presented in this report. 1.5 Description of the Bayou La Batre WWTP The new Bayou La Batre (Buford Bryant) WWTP is located at Railroad Street in Bayou La Batre, AL. It came online in 2012 and replaces the old WWTP located nearby off Shell Belt Road. The average daily flow at the Bayou La Batre is 1.0 million gallons a day (MGD). The design flow for the facility is 3.0 MGD, which is also the permitted flow, with the capability of handling a peak flow of 9.0 MGD. The WWTP uses tertiary treatment and sand filtration with ultraviolet (UV) disinfection. It also includes a three cell aerobic digester for sludge handling and a belt filter press for sludge dewatering. The UV system has two banks a primary and a back-up and UV bulbs are changed on a routine maintenance schedule. A portion of the treated water is re-used at the WWTP. The WWTP has a new SCADA system and a number of back-up systems in the event of a failure, including back-up generators. Influent at the Bayou La Batre WWTP consists of domestic sewage and wastewater from nearby seafood processing plants, particularly the Sea Pearl plant. The Sea Pearl plant is a major contributor to the current 1.0 MGD average daily flow at the WWTP, and flows can decrease significantly when the Sea Pearl plant is not operating. There are about 750 customers on new sewer lines. The new outfall/diffuser for the WWTP is located at N, W and is shown on the map in Figure 1. It s located about a mile southwest of the old outfall. The diffuser has eight 4 wide ports. Treated effluent is pumped about 2 miles from the WWTP to the diffuser and has a long residence time in the outfall pipe. 1.6 General Description of Study Design Prior to the main dye study on March 10-13, 2014, a preliminary dye study was conducted on March 7, 2014 to estimate the residence time of the effluent in the outfall pipe and see if the dye was impacted more by the direction of the tides or the direction of the winds. Drogues were also deployed on March 7, 2014 to try and determine the timing of the turning of the tide and ascertain the influence of winds. In addition to the drogue study, background testing of Portersville Bay and Grand Bay was conducted with two boat tracking fluorometers at the surface and a profiler fluorometer at depth. Six stations equipped with WET Labs fluorometers (WET Labs, Inc., Philomath, OR) to measure dye concentrations and Star-Oddi CTDs (Star-Oddi, Ltd, Iceland) to monitor conductivity, temperature, and depth were deployed at strategic locations to the east and west of the diffuser on March 7, Figure 1 shows a map of the study area with the six station locations, the old outfall, and the new diffuser, as well as the shellfish growing area classifications. The stations 7

8 remained in the water until March 13, 2014, except for Station 5 which was retrieved on March 12, 2014 due to stormy weather conditions in the area of its deployment. The dye for the comprehensive study was injected over a full tidal cycle (24.8 hours) from 3:10 AM on March 10, 2014 to 3:58 AM on March 11, 2014 and remained in the system for at least four days. Boat tracking with towed WET Labs fluorometers was conducted to find the edges of the dye plume during daylight hours and the surface dye concentrations, in addition to the 24/7 dye readings recorded by the station-attached submersible WET Labs fluorometers. In addition to the dye study, FDA collected samples of influent, pre-uv treated effluent, and post-uv treated effluent from the WWTP. Samples were analyzed for FC, MSC, AdV, and NoV GI and GII. These analyses were performed to assess the efficiency of the WWTP s tertiary treatment process and UV disinfection at removing viruses and to better inform FDA s recommendations for sizing a prohibited area around the WWTP diffuser. 1.0 METHODS - See Appendix 1 for detailed methods used in the study. 3.0 RESULTS 3.1 Drogue Study and Preliminary Dye Study The drogues were released shortly after the tide turned from flood to ebb. The drogues were pushed by the wind and moved northeast even though the tides were moving south on an ebb tide (although the tides appeared to be at slack based on visual observations). The results of the drogue study indicated that winds were the predominant factor, which was confirmed by the preliminary dye study. The exact timing of the turning of the tide couldn t be determined based on the drogue study. During the preliminary dye study, dye was first seen at the outfall at 3:11 PM. Since the injection began at 1:20 PM, the travel time through the outfall pipe was 1.7 hours. Using the results of the preliminary dye study and tidal charts for Bayou La Batre, FDA determined that 3:10 AM on March 10 th (1.7 hours before slack ebb) would be the best time to start the dye injection. 3.2 Weather Conditions According to NOAA s National Weather Service ( and the Garland Pump rainfall station in Bayou La Batre, no precipitation fell during the dye injection period. The Garland Pump station recorded a total of 0.45 inches of rainfall in the afternoon on the second day of the study, March 11, 2014, but no precipitation occurred on the third or fourth days of the study. Therefore, rainfall was not a significant factor during the dye injection or the overall study period. As observed during the preliminary dye study, winds influenced the direction of dye movement, with dye frequently moving in the direction the winds pushed it during the study period, regardless of the tides. Maximum wind speeds and directions were as follows: 17 mph SW on March 10 th, 18 mph SE on March 11 th, 26 mph N on March 12 th, and 21 mph N on March 13 th. 8

9 3.3 Dye Injection Records from the WWTP showed that the flow during the dye injection period was 0.75 MGD, which is lower than the average daily flows at the plant (approximately 1.0 MGD when the Sea Pearl seafood plant is operating). FDA determined using a mass balance approach that the dye concentration flowing out of the WWTP was 1345 ppb. 3.4 Travel Time This study determined the extent of dye travel on the flood tide and the ebb tide on the first day of the study (March 10, 2014) and on successive tides. As seen in Figure 2, the dye velocity was determined to be 0.70 mph. This was based on dye readings at the leading edge of the dye plume around 3:00 PM on March 10 th, which was just east of Bayou Coden, and on the average velocity of dye travel to the stations and random points along the plume. Based on a velocity of 0.70 mph and the distances to the aquaculture sites, it was determined that travel time of the dyetagged effluent to the aquaculture sites near Fowl River would be 7.1 hours and to the aquaculture sites in Grand Bay would be 4.3 hours. On the first day of the study, winds were out of the southeast and moving at speeds up to 18 mph. The winds mainly pushed the dye northwest towards the Fowl River aquaculture sites on that day. However, it was assumed that winds from the southeast could push effluent at the same velocity (0.70 mph) towards Grand Bay in the northwest should the wind direction shift. Under higher wind velocities, the dye-tagged effluent could move even faster towards the aquaculture sites. The prohibited area radius around the diffuser for a 400:1 dilution zone, as described later in this report, is 1.43 miles for the current average daily WWTP flow of 1.0 MGD. For a 1.43 mile radius, effluent moving at a velocity of 0.70 mph would take 2.1 hours to reach the boundary of the 400:1 dilution zone. Therefore, the recommended response time to a failure event at the Bayou La Batre WWTP is less than 2.1 hours in terms of closing the conditionally approved growing area. If the aquaculture sites are the only portions of the area being harvested at the time of a failure event, more response time may be available for closing those sites. 3.5 Dye Readings at Stations Dye readings recorded by the station WET Labs fluorometers and boat tracking fluorometers within a 100 meter radius of each station are shown in Figures 3-7. The WET Labs fluorometer at Station 3 malfunctioned and did not collect any dye data so no results are shown for that station. Tidal depth in feet is also plotted based on the Star-Oddi CTD readings at Station 2. The CTDs at the other stations malfunctioned and did not record depth data, but the CTD data at Station 2 is representative of tidal cycles in the area if not the exact depths at each station. Most depths throughout Area II were in the range of about 3 6 feet. Accumulated peak 1 hour average concentrations for each 12 hour period of the study are plotted in Figures 3-7. Because flows out of the diffuser were inconsistent and the movement of the dye was highly affected by wind-driven circulation, accurate steady state dilution levels could not be determined using the superposition method. For example, at Station 1 (Figure 3) dye was not detected by the station fluorometer on March 11 th, but it was detected again on March 12 th. It 9

10 is also possible that dye remained in the growing area after March 13, 2014, when the stations were pulled from the water due to high winds and predicted storms for the following days. If dye did remain in the area on March 14 th and possibly March 15 th, the dye levels are unknown and cannot be considered as part of a steady state analysis for the stations. However, by adding the peak 1 hour average dye levels for each 24 hour period of the study together, the accumulated dye concentration value and associated dilution value provides a good reference point for how much dye was reaching the station over the study time period, even if the result is not fully representative of the steady state condition. The dye readings recorded by the boat tracking fluorometers at the surface within a 100 meter radius of each station are also shown in Figures 3 7. The values from each day were accumulated to determine how much dye reached the surface waters near the station during the overall study period. This accumulated value was then compared with the steady state peak 1 hour average value from the station fluorometer on the bottom. With the exception of Stations 4 and 6 (see Figures 5 and 7), where the surface dilution was estimated to be 740:1 and the bottom dilution was estimated to be >100,000:1 for Station 4 and the surface dilution was estimated to be 1,041:1 and the bottom dilution was estimated to be 2880:1 for Station 6, results at the surface and the bottom did not differ that much. This was confirmed by profiles conducted from the top of the water column to the bottom near each station. Due to the low depths in the area, significant differences in dye concentrations from the top of the water column to the bottom were not observed. As expected, the stations closest to the diffuser Station 1 to the east and Station 4 to the west had the lowest dilution levels of the dye-tagged effluent. The minimum dilution at Station 1 was 708:1 (Figure 3) and the minimum dilution at Station 4 was 740:1 (Figure 5), based on accumulated surface dye readings. Station 2 had the highest dilution level at 2474:1, even though it was closer in distance to the diffuser than Stations 5 or 6. However, Station 2 was over a mile south of the diffuser, and it was observed during the study that dye levels tended to decrease to the south. Station 5, which was close to the aquaculture sites in Grand Bay, had a bottom dilution level of 854:1 and a surface dilution level of 1388:1 (Figure 6). Dilution levels at Station 5 were lower than expected, given that the station was over 3 miles from the diffuser. There was no station located at the aquaculture sites near Fowl River, but Station 6 was the closest to those sites and had a minimum dilution level of 1041:1 based on accumulated readings at the surface. 3.6 Dye Readings by Tracking Fluorometers Figure 8 represents the accumulated 5-point moving average concentration values and the corresponding dilution levels for the four days of the study period (March 10 13, 2014) as determined by both WET Labs boat tracking fluorometers. The raw data used to create these figures can be provided in Excel worksheets upon request. The minimum dilution found via the boat tracking fluorometers on March 10 th was 19:1, equivalent to a 5-point moving average concentration of 70.5 ppb and shown in Appendix 3, Figure 9. This level was found at 0.01 miles from the diffuser. Therefore, the initial minimum dilution of the dye-tagged effluent as it exited the diffuser was 19:1. This is a relatively low level of dilution for a diffuser, but the depth of water above the diffuser was only 4 feet at the time the dye readings were taken. 10

11 As previously noted and shown in Figure 2, the dye-tagged effluent was pushed strongly to the west on March 10 th at a velocity of 0.70 mph. Dye was not detected to the west of the outfall on the first day of the study, but was detected at significant levels (>1.0 ppb) to the west near Station 4 on the second day of the study, representing dilution levels of around 1345:1. Dye was not detected at the aquaculture sites near Fowl River via the WET Labs FLRHRT 2487 boat-tracking fluorometer on the first day of the study because tracking with that fluorometer ended around 3:00 PM. However, based on the velocity and trajectory of the dye, it s expected that dye reached those aquaculture sites around 5:00 PM on the first day. Dye levels detected near the aquaculture sites on the second, third, and fourth days of the study were in the range of ppb, ppb, and ppb, respectively. The accumulated dye totals at the Fowl River aquaculture sites were ppb, equivalent to dilutions of :1. Since dye levels on the third day were higher than on the second day, there appeared to be a build-up of dye at the aquaculture sites near Fowl River. However, there was a southeasterly wind on the second day, which pushed the dye northwest, away from those sites. The winds on the third day were westerly winds, so it is reasonable that dye levels at the Fowl River aquaculture sites were higher on the third day than on the second day based on the significant impact of wind direction. As shown in Figure 8, accumulated dye concentrations near Station 5 and the aquaculture sites in Grand Bay were mainly in the range of ppb, equivalent to dilutions of :1. When the highest boat tracking readings at the surface were accumulated, the equivalent dilution was 1388:1. As shown in Figure 6, the bottom dilution level determined using the Station 5 fluorometer data was 1232:1, but Station 5 was pulled a day earlier than the other stations. The accumulated dye concentration recorded by the station fluorometer would have been higher (and the dilution lower) had the fluorometer been left in the water on March 13 th, since dye was detected on the surface in Grand Bay at levels up to 0.30 ppb on that day. Surface dye levels closer to 0.10 ppb were found near the station. Since dye was still detectable at significant levels ( ppb in many parts of Area II) on the final day of boat tracking, March 13, 2014, it s likely that dye would have also been detectable on March 14 th and possibly even March 15 th or 16 th had tracking of the dye been able to continue and had the station fluorometers been able to stay in the water. Adverse weather conditions placed the instruments at risk and due to the expense of the instruments, a decision was made to retrieve the fluorometers and CTDs on March 13 th. However, based on the time period of when the stationed fluorometers were collecting data dye tag effluent was still detected at Stations 2 and 5 indicating the overall residence time of the dye tagged effluent is greater than 2 tidal days. When accumulating the dye levels on each tidal day the accumulated levels appear to be still building indicating that pollutants discharged at a continuous rate may take longer than 2 tidal days to reach a steady state maximum. And subsequently, pollutants will take greater than 2 tidal days to be completely flushed from the study area. 11

12 3.7 Profiles of Dye at Depth Several profiles were conducted using the SeaBird CTD interfaced with a WET Labs tracking fluorometer (WET Labs FLRHRT 586). Due to the low depths in Portersville Bay and Grand Bay (4 8 feet or less), dye levels at the bottom were the same as at the surface no stratification of the dye was observed. 3.8 Projections for Different Wastewater Treatment Plant Flows The dye injection was conducted on March 10, 2014 during a relatively low wastewater treatment plant flow period. The average flow recorded by the SCADA system during the dye injection period was 0.75 MGD, compared to a normal average daily flow of 1.0 MGD. The WWTP operators reported that the Sea Pearl seafood processing plant, the largest contributor to the WWTP flows, was not operating on the day of the dye injection, which is likely the reason for the lower flow value. Also, the WWTP operators reported that new flow sources will be added to the WWTP collection system in the future, since the design capacity of the plant is 3.0 MGD. The calculated dilution values of the effluent would be lower under high flows than low flows. Therefore, projections for 1.0 MGD, 2.0 MGD, and 3.0 MGD flows were modeled in ArcGIS based on the data collected under 0.75 MGD flows. These projections are shown in Figures Figure 9 shows the accumulated dye levels and the 400:1 and 1000:1 dilution zones for a projected flow of 1.0 MGD. Since the current average daily flow at the WWTP is 1.0 MGD, it is recommended that the map in this figure be used in sizing the current prohibited area around the new diffuser. The use of 400:1 vs. 1000:1 dilution will be discussed more below in Section Dilution levels near the Fowl River aquaculture sites were greater than 1000:1 but less than 1345:1 under the projected 1.0 MGD flow. Figures 11 and 12 show the dilution zones and accumulated dye tracking results for projected flows of 2.0 MGD and 3.0 MGD, respectively. Under projected flows of 2.0 MGD, the 400:1 dilution radius would still be only 2.0 miles from the diffuser, but the minimum dilutions at the aquaculture sites near Fowl River would be around 560:1, which could be a concern if viruses are present in the effluent under the higher flow conditions. Figure 11 shows that under projected flows of 3.0 MGD, dilution levels at the Grand Bay aquaculture sites would be less than 400:1 and at the Fowl River aquaculture sites would be less than 1000:1. For this reason, FDA does not recommend that the growing area remain conditionally approved under flows of 3.0 MGD unless a significant amount of microbial data for the influent, pre-uv effluent, and post-uv effluent can be provided to support the use of a lower level of dilution and the growing area can be closed in a very short period of time in the event of a failure at the WWTP. Under flows of MGD, a response time of less than 2.1 hours would likely be needed to close the conditional growing area before sewage reaches it, but under flows of 2.0 MGD 3.0 MGD, an even faster response time (< 1 hour) would likely be needed. 12

13 3.9 Microbiological Analysis of WWTP Influent and Effluent Table 1 shows the fecal coliform (FC) and male-specific coliphage (MSC) levels in the influent at the Bayou La Batre WWTP. FC were present at 5.4 x 10 6 cfu/100 ml and MSC were present at 3200 pfu/100 ml in the influent. Tables 2 and 3 show the levels of FC and MSC in the pre-uv treated and post-uv treated effluent, respectively. The WWTP was efficient at removing both FC and MSC. The reduction of FC in the pre-uv treated effluent was 4 logs and in the post-uv treated effluent was about 1 2 logs. MSC levels were 40 pfu/100 ml in the pre-disinfected effluent and 10 pfu/100 ml in the post-disinfected effluent. Table 4 shows the norovirus (NoV) GI and GII and adenovirus (AdV) results for the influent, pre-uv treated effluent, and post-uv treated effluent. NoV GI was detected at a level of 1.17 x 10 2 and AdV was detected at a level of 1.38 x 10 3 RTqPCR units/100 ml in the influent, but NoV GII was not detected in the influent. NoV GI and NoV GII were not detected in the pre-uv treated effluent or post-uv treated effluent either. AdV levels in the post UV-treated effluent were 1 2 logs lower than in the influent. Because NoV GII was not detected in the influent, the WWTP s removal efficiency for NoV GII could not be determined based solely on the results of this study. However, the WWTP was efficient at removing NoV GI, AdV, FC, and MSC Short Term Failure - Dilution and Anticipated Fecal Coliform (FC) Concentrations A short-term raw sewage failure at the Bayou La Batre WWTP is unlikely given the WWTP s advanced SCADA system, new technologies, and multiple back-up systems. A short-term UV disinfection failure is also unlikely, particularly given the routine maintenance of the bulbs and the back-up bank, but is possible due to bulb malfunctions and other potential issues with the UV system. However, both failure events are assessed below and should be considered as part of the conditional management plan for the area. In the unlikely event of a raw sewage failure, a dilution of :1 would be needed for any approved areas within an accessible distance of the diffuser, since FC levels in the influent were 5.40 x 10 6 cfu/100 ml. A :1 dilution line could not be determined based on the results of this study because dye diluted to that level would be below the limit of detection of the fluorometers (0.03 ppb in estuary water). However, a dilution versus distance relationship was used to estimate the impact of a failure based on the maximum surface dye levels found during the study in the far-field region beyond the outfall as shown in Appendix 3, Figures 5 through 8. Based on the regressions if a loss of disinfection occurred a larger area beyond the minimum sized prohibited zone would be needed. Assuming that a :1 dilution would be sufficient to dilute a raw sewage failure the buffer zone would need to increase to miles based on the regressions for a WWTP flow rate of 0.75 MGD and potentially a larger area for flows >0.75MGD. The highest level of FC detected in the effluent was FC/100 ml. In order to dilute that to the approved area standard of 14 FC/100 ml, a dilution of 929:1 would be needed. Therefore, a 929:1 dilution zone would be sufficient to reduce the FC from a UV disinfection failure event to 13

14 the approved area standard under the conditions observed during this study. Nevertheless, since flows may be higher during a failure event, wind conditions may be worse, pre UV-treated FC levels could be higher than observed in this study, and many other factors could impact the quality of the effluent discharged and conditions within the growing area. Thus, using a more conservative value and assuming that 10000:1 dilution would be sufficient to dilute a loss of disinfection failure the prohibited zone would need to increase to a radius of at least 4.11 miles based on the regressions for a WWTP discharging at a daily average flow rate of 0.75 MGD considering the first ebb tide data as shown in Appendix 3 Figure 7. However, when the far-field data is considered from two days after the dye release, the prohibited zone would need to increase to 6.76 miles. The table below indicates the size of the prohibitive zone estimated for the various flow rates considered under failure (loss of disinfection) conditions. Flow Rate (MGD) Dilution needed for loss of disinfection Buffer Zone Size (miles) : Estimates where not determined for the size of a prohibited zone for a raw sewage release as the estimates fall beyond the range of data collected, however, the distance versus dilution regressions found in Appendix 3 suggests that an emergency closure area would have to extend greater than 10 miles. FDA recommends that the entire growing area be closed as quickly as possible in the event of either a raw sewage release or a UV disinfection failure at the WWTP Determination of 400:1 and 1000:1 Dilution Under Scenario 2 for sizing prohibited areas (see Section 1.2), the size of the prohibited zone can be reduced and a conditional area can be established if a 1000:1 dilution zone is achieved and other conditions are met. For WWTPs with UV disinfection and efficient virus removal, such as the Bayou La Batre WWTP, a smaller 400:1 dilution zone can be established around the diffuser. Figure 9 shows the 400:1 dilution zone and 1000:1 dilution zone based on radii around the diffuser using the maximum accumulated dye concentrations at those radial distances for a projected flow of 1.0 MGD. The 400:1 dilution radius is 1.49 miles from the diffuser and the 1000:1 radius is 1.94 miles from the diffuser. The 400:1 dilution zone in this figure can be used to establish a prohibited area around the diffuser because viruses and viral indicators in the WWTP s effluent were very low. If future microbial testing should indicate a problem with viruses in the effluent, e.g., as WWTP flows increase, FDA recommends use of a 1000:1 dilution zone in place of the 400:1 dilution zone. However, if average daily flows at the WWTP increase to 2.0 MGD, as shown in Figure 10, the 1000:1 dilution zone will encompass both the aquaculture sites in Grand Bay and near Fowl River. If average daily flows at the WWTP increase to 3.0 MGD, as shown in Figure 11, the 400:1 dilution zone will also encompass all of the aquaculture sites. For this reason, the impact 14

15 of any significant increases in average daily WWTP flows in the future should be considered in relation to the impact on classifying and managing the growing area. 4.0 CONCLUSIONS AND RECOMMENDATIONS When considered collectively, the data from the hydrographic dye study at the Bayou La Batre WWTP and the microbiological assessments of WWTP influent and effluent support the following conclusions and recommendations: The WWTP is very efficient at removing FC indicator bacteria and in meeting its permitted requirements for FC. The WWTP is also efficient at removing MSC, AdV, and NoV GI. MSC and NoV GI were not detectable in the post-uv treated effluent. NoV GII was not consistently detected in the influent or effluent during the study period. Thus, the WWTP s ability to remove NoV GII could not be assessed. For a projected flow of 1.0 MGD, the average daily flow at the WWTP, the 1000:1 dilution zone as determined by boat tracking of dye at the surface was achieved at a radius of 2.01 miles from the diffuser, and the 400:1 dilution zone was achieved at a radius of 1.43 miles from the outfall (see Figure 9). FDA recommends establishing a 400:1 dilution zone around the diffuser as shown in Figure 9, since the Bayou La Batre WWTP uses UV disinfection and the post-uv treated effluent from the plant does not contain significant levels of viruses or viral indicators. However, if future microbial testing demonstrates a concern for viruses in the final effluent, e.g. as flows and domestic sewage inputs increase, FDA recommends employing 1000:1 dilution rather than 400:1 dilution in sizing the prohibited area. Wind-driven circulation had a much greater impact on the direction the dye-tagged effluent traveled than the slow moving tides in the area. Winds moved the dye at a relatively high velocity (0.70 mph) towards the aquaculture sites to the east on the first day of the study. Dye-tagged effluent remained detectable near the growing area for at least 4 days (March 10 13, 2014) and relatively high levels of dye ( ppb) were observed near the Fowl River aquaculture sites on the third day of the study. FDA was unable to conduct dye tracking on March 14-16, 2014, but it is possible that dye remained in the system on these days and that it takes at least 6 days for effluent to be flushed from the system by the tides. The aquaculture sites near Fowl River are likely to receive high levels of effluent in the event average daily flows at the WWTP increase significantly. Estimated travel time of the effluent from the diffuser to the aquaculture sites near Fowl River is approximately 7.14 hours. Estimated travel time of the effluent to the aquaculture sites in Grand Bay is approximately 4.29 hours. Travel time to either of the sites could be even shorter if strong winds push the effluent at faster speeds towards the sites. Estimated travel time to the border of a prohibited area based on a 400:1 dilution zone with a 1.43 mile radius is 2.1 hours. ADPH noted that it can respond to a failure event at the WWTP in 12 hours in order to close the growing area, but based on the fast travel times determined in this study (due in large part to strong winds), it is recommended that ADPH have the ability to close the conditionally approved 15

16 area in less than 2.1 hours in the event of a raw sewage failure or UV disinfection failure at the WWTP. The advanced monitoring systems at the WWTP and good communications procedures between the WWTP operators and ADPH should assist in this effort. FDA recommends that a documented communication system be set up between ADPH, ADCNR, the Bayou La Batre WWTP, and the licensed harvesters of the growing area to ensure that harvesters are notified to stop harvesting in a sufficient period of time based on the location of the harvest sites in the unlikely event of a failure at the WWTP. If no harvesting takes place immediately after a failure event, this will allow the ADPH additional time to close the growing area. Based on the microbial findings in samples from the WWTP and the results of the dye study, FDA recommends establishing a prohibited area around the diffuser with a level of 400:1 dilution at a 1.43 mile radius or 1000:1 dilution at a 2.01 radius for flows of 1.0 MGD as shown in Figure 9. If flows increase to 2.0 MGD, FDA recommends expanding the prohibited area to at least a 2.00 mile radius from the diffuser as shown in Figure 10 considering a level of dilution of 400:1. However, flows of 3.0 MGD or greater could result in less than 400:1 dilution at the aquaculture sites near both Grand Bay and Fowl River, as shown in Figure 11, so additional microbial data from the WWTP would be needed at flows that high in order to support maintaining the conditionally approved designation of the growing area. FDA plans to continue conducting MSC and virus testing of samples from the WWTP. If future testing shows that the efficiency of the WWTP to remove viruses decreases as flows increase or the WWTP becomes older, FDA would recommend using 1000:1 dilution instead of 400:1 dilution in sizing the prohibited area around the diffuser to increase the dilution of viral loads. 16

17 References: Alabama Department of Public Health Seafood Branch Shellfish Growing Water Report for Area II. Division of Food, Milk, & Lodging, Mobile, AL. Burkhardt, W., III., J. W. Woods, J. Nordstrom, and G. Hartman A real-time RT-PCR protocol for the simultaneous detection of norovirus and enteroviruses. U.S. Food and Drug Administration, Washington, DC. Cabelli, V. J Microbial indicator levels in shellfish, water and sediments from the upper Narragansett Bay conditional shellfish-growing area. Report to the Narragansett Bay Project. Narragansett Bay Project, Providence, RI. DePaola, A., J.L. Jones, J. Woods, W. Burkhardt III, K.R. Calci, J.A. Krantz, J.C. Bowers, K. Kasturi, R.H. Byars, E. Jacobs, D. Williams-Hill, & K. Nabe Bacterial and Viral Pathogens in Live Oysters: 2007 United States Market Survey. Appl. Environ. Microbiol. 76(9): Kilpatrick, F.A Techniques of Water-Resources Investigations of the United States Geological Survey: Simulation of Soluble Waste Transport and Buildup in Surface Waters using Tracers. United States Geological Survey, Chapter A20, Book 3, Application of Hydraulics, Report Number: TWRI3A20. National Shellfish Sanitation Program Revision. Guide for the Control of Molluscan Shellfish. Interstate Shellfish Sanitation Conference and U.S. Food and Drug Administration pdf 17

18 APPENDIX 1 METHODS (Section 2.0) 2.1 Dye Standard Preparation and Fluorometer Calibration The dye tracer used in this study was Rhodamine WT, purchased from the Keystone Aniline Corporation, with a specific gravity of approximately 1.12 (20% as dry dye). Ten (10) standards were prepared from the stock solution of Rhodamine WT dye and distilled water by serial dilution, ranging from 100,000 parts per million (ppm) to 0.1 parts per billion (ppb). The Rhodamine WT dye was detected and its concentrations in Portersville Bay and Grand Bay were obtained using a combined total of nine fluorometers. Six of these were WET Labs FLRHB submersible fluorometers (WET Labs, Inc., Philomath, OR) that were attached to the six stations deployed in the growing area throughout the course of the study. Two were WET Labs FLRHRT fluorometers that were pulled behind a boat and used for tracking the dye on each day of the study. The final was a WET Labs FLRHRT fluorometer interfaced with a SeaBird SBE19-plusV2 CTD used for conducting profiles of the dye at depth while at the same time capturing conductivity, temperature, and depth data within the water column. The dye standards were used to develop calibration curves for FDA s WET Labs FLRHRT 2040, 2487, and 586 tracking and profiling fluorometers and the six station fluorometers WET Labs FLRHB units 913, 915, 1730, 1731, 2032, and With the subtraction of background fluorescence levels in the bay, these curves were used to calculate part per billion (ppb) levels of dye based on the WET Labs measured fluorescence units (FUs). The y-intercept of the calibration curve was adjusted so that a 0.1 ppb result read as a perfect 0.1 on the curve. The slope and x-axis values for the curve remained the same, but this adjustment caused a slight addition of error (5-10% error) to the higher concentrations on the curve, such as 10 ppb and 100 ppb. However, higher accuracy at the lower end of the curve, 0.1 ppb, is more vital in order to optimize sensitivity in detecting the dye at low concentrations, as important data tends to fall within the ppb range during FDA dye studies. Using a calibration curve adjusted in this manner is necessary when converting raw FU readings to ppb values if sensitivity in the ppb range is critical for the study. The WET Labs limit of detection in distilled water is 0.01 ppb, with a limit of detection in estuary water of approximately ppb dependent on the specific fluorometer. Background readings were captured three days prior to the study on March 7, For the interfaced SeaBird CTD and WET Labs FLRHRT 586 fluorometer (a.k.a., the profiler ), background levels were recorded in terms of voltage readings and were converted to ppb units by applying a conversion factor and calibration curve data. However, the average of the raw voltage readings was used to program the background level for the profiler in RAFT-MAP. Background levels for the station fluorometers were determined by plotting all of the data collected by the fluorometers and finding the baseline FU level for readings taken prior to the dye injection in comparison with those recorded after the dye injection at each station. Background levels for the tracking fluorometers were based on maximum FU readings in the growing area, excluding outliers, detected prior to the dye injection. These background levels were programmed into

19 RAFT-MAP and automatically subtracted from the fluorescence readings recorded in the bay after the dye injection. 2.2 Drogue Study and Preliminary Dye Injection Orange and grapefruit drogues were used on March 7, 2014 to assess the timing of tidal cycles (i.e., slack high/start of ebb tide) and to assess the impact of winds vs. tides on drogue movement. The drogues were released near the diffuser shortly before the predicted turning of the tide from flood to ebb at 2:34 PM and were tracked via boat. In addition to the drogue study, a small dye release was conducted to find the travel time of effluent through the WWTP outfall pipe. The outfall pipe is over 2 miles long, but the dye injection took place at a U-pipe approximately 1 mile from the diffuser that is used to prevent backwash of estuary water into the pipe rather than at the WWTP. The preliminary dye injection started at 1:20 PM. Approximately 1.5 gallons (5000 ml) of a dye/water mixture (3000 ml of dye, 2000 ml of water) was injected at a rate of 63 ml/min with a pump speed of 300 RPM for about 1.3 hours. A boat was stationed near the diffuser to determine the time at which the dye first exited the diffuser and could be seen and detected by fluorometers. The results of the preliminary dye injection were used to determine the start time for the dye injection on March 10, Dye Injection For the dye injection, a total of 9.5 gallons of dye was injected at a constant rate into the WWTP effluent over a 24.8 hour period from 3:10 AM on March 10, 2014 to 3:58 AM on March 11, To facilitate the pumping of dye, 5 gallons of deionized water was added to 7 gallons of dye creating a 40:60 water/dye dilution mixture (12 gallons total, of which 9.5 gallons was injected). A Masterflex model variable speed peristaltic pump (Cole-Palmer Instrument Co.) was used to withdraw the tracer dye solution from a large plastic holding bin, using Masterflex Tygon L/S-14 tubing. A pump head size 7014 was used with a constant pumping rate of 25 ml/min which was maintained at about 101 revolutions/minute (rpm) head speed. The tracer dye mixture was fed continuously into the final effluent following the UV treatment over the 24.8 hour injection period. The dye tubing became clogged for a 45 minute period around 7:30 AM on March 10, 2014 before the clog was discovered and removed, so this reduced the amount of dye injected from 9.8 gallons, the amount that should have been injected at a pump rate of 25 ml/min for 24.8 days, to 9.5 gallons. The dye was injected at the same U-pipe 1 mile from the diffuser as was done during the preliminary dye injection. The initial concentration of the dye in the effluent was determined using the WWTP s flow average over the period of the dye injection (0.75 MGD). Flow rates out of the WWTP were based on SCADA readings. 2.4 Dye Tracking Boat tracking was conducted on each day of the study with two boat-towed fluorometers, the WET Labs FLRHRT-2040 and WET Labs FLRHRT-2487, to track the dye past the cages; to determine the shape and edges of the dye plume; and to assess the dye concentrations and dilutions in the surface waters. The fluorometers were linked to Panasonic Toughbook C-19

20 field computers operating FDA s custom-made mobile GIS software, RAFT-MAP (Real-Time Application for Tracking and Mapping). Two boats were used, with each instrument on a different boat. Dye readings were also taken on successive days (March 11-13) for high and low tides. Traverses were done on all the days of the study from north to south and east to west and vice versa, and dye readings were also recorded at each of the fixed station locations to show changes in dye concentration and build-up with time. While traverses of the dye were being done with two of the WET Labs FLRHRT fluorometers, the other FLRHRT fluorometer (586) was interfaced with a SeaBird SBE19-plusV2 CTD used for conducting profiles of the dye at depth at various locations along the path of the dye plume, particularly near the diffuser and each of the station locations. Fluorescence data from the SeaBird interfaced with the WET Labs was transmitted in voltage readings, but these were later converted to ppb readings using the dye calibration data. A five-point moving average was applied to the dye concentration data to smooth out any false high or low readings. Dilution was calculated by dividing the initial concentration of dye injected at the WWTP by the final (five-point moving average) concentrations in the estuary. Using RAFT-MAP, the fluorometer dye concentration readings (in FUs) with the associated GPS readings were converted into ppb units and automatically plotted on a field GIS map in real-time on the boat. The GIS caches were later synchronized into ArcGIS Desktop to post-process the data (e.g., remove false positive readings); add scales, legends, station locations, growing area classification lines, and other map features; and provide additional information, such as the accumulated dye concentrations and locations of dye readings with 1000:1 and 400:1 dilution. The velocity of dye movement was determined by dividing the distance the leading edge of the dye traveled from the diffuser at 10:00 AM to east of Bayou Coden at 2:54 PM on March 10, 2014 by the time of travel (about 5 hours). The velocity calculated in this manner was similar to the average velocities of dye travel from the diffuser to the stations and from the diffuser to random points along the dye plume. Using the maximum velocity values produced results that were unrealistic and did not much up with the location of the leading edge of the dye, whereas the average velocity values did. Therefore, the velocity based on average values and the location of the leading edge of the dye was used. A map was created in ArcGIS Desktop of estimated travel times from the diffuser to various locations, including the aquaculture sites near Fowl River and Grand Bay, based on the velocity of dye movement. The Geostatistical Tool in ArcGIS Desktop was used to create the dye contour maps interpolating the data and estimate dye concentrations in areas where no dye tracking was conducted based on the surrounding areas where dye tracking was conducted. This tool was also used to add the actual and interpolated dye concentrations for each day of the study, March 10 13, 2014, to achieve accumulated total dye concentrations throughout the growing area for all four days of the study. The result was mapped in ArcGIS as the total dye mass for a 0.75 MGD flow (the flow rate during the study) over the March 10 13, 2014 study period.

21 Projections were also conducted for what the dye mass would look like based on elevated flows of 1.0 MGD (the current average daily flow at the plant), 2.0 MGD, and 3.0 MGD for the March 10 13, 2014 study period using the Geostatistical Tool in ArcGIS Desktop. Since the WWTP has a design flow of 3.0 MGD, this was the maximum projected flow level that was modeled. The radius estimates based on flow rate were determine from the actual data collected under the 0.75 MGD flow rate and estimating dilutions at these data points under various flow rate conditions. Appendix 3, Figures 1-4 display field data points collected that are represented under various flow rates and color coded to display values that fall below 400:1 dilution (data points displayed in blue) as well as values that fall below 1000:1 (data points displayed in purple). 2.5 Dilution Analysis - Dye Readings from Station Fluorometers One of the advantages of the station fluorometers over the boat-towed fluorometers is that they can detect dye every ten minutes for thirty second intervals over the entire four day cycle of the study and can therefore pick up dye readings at depth during hours in which boat tracking was not possible. The fluorescence readings recorded by the submersible fluorometers at each of the six stations were downloaded, converted to ppb using each fluorometer s calibration curve chart, and plotted in SigmaPlot alongside the Star-Oddi CTD tidal depth curves for the study period. The accumulated boat-towed fluorometer readings at the surface within a 100 meter radius of each station were included on the charts as well. A five-point moving average was applied to the dye concentration data to normalize high or low readings in the data. Dilution was calculated by dividing the initial concentration of dye injected at the WWTP by the final (five-point moving average) concentrations detected in the estuary. Since only a 24.8 hour dye injection was conducted, FDA attempted to use the superposition method (Kirkpatrick, 1993) to estimate the steady state condition for dye at each of the stations using data collected from March 10 13, FDA has successfully employed the superposition method in a number of recent studies and uses this method to save time and resources. However, in the case of Bayou La Batre, because flows out of the diffuser were inconsistent (frequent stops and starts); the movement of the dye was highly affected by winddriven circulation; and the stations had to be pulled early on March 13, 2014 due to high winds and impending storms (even though dye may still have been detectable on March 14 15, 2014), accurate steady state dilution levels could not be determined using the superposition method. For these reasons, use of the superposition method to add the dye concentrations for each tidal cycle at each station may not adequately represent steady state concentrations and dilutions. However, by adding the dye levels for each 24 hour period of the study together, the accumulated dye concentration value and associated dilution value provides a good reference point for how much dye was reaching the station over the entire study period, even if the result is not fully representative of the steady state condition. Peak 1 hour average concentrations of dye at each station were determined. Maximum and average concentrations were also determined but were not plotted because FDA has found that these values are overly conservative or not conservative enough in representing the higher dye levels reaching the station. Peak 1 hour average values were determined by plotting the dye data

22 for each 24.8 hour period (ebb tide or flood tide) at each station and finding the 1 hour period with the highest average dye concentration value within that period. FDA found in past studies and in this study that in some cases the maximum dye concentrations were overly conservative because they included outlier values, so the peak 1 hour average concentrations and dilutions are given more weight than the maximum concentrations and dilutions. Finding the hour during each tidal cycle in which the station received the highest amount of dye and averaging the dye concentration readings over that 1 hour period gives dye concentration values and steady state dilution estimates that are both realistic and conservative. First the peak 1 hour average values for the two 24.8 hour tide periods on the day of the dye injection, March 10, 2014, were plotted. For the second day of the study, March 11, 2014, dye still remained in the system, so the peak 1 hour average values for this remaining dye level were added to the values detected on day 1 and plotted. Remaining dye levels on successive tides on March 12-13, 2014 were also added. If the superposition principle were used, a steady state condition would be achieved when all the dye levels were added together and no dye remained detectable near the stations. However, most of the stations were still detecting some dye at the time they were pulled from the water on March 13, Station 5 was retrieved on March 12, 2014, a day earlier than the other stations, to protect the fluorometer and CTD from stormy weather conditions predicted for Grand Bay that evening. For each station, the minimum dilution based on either the peak 1 hour average concentrations from the station fluorometer or the maximum concentration detected by the boat-tracking fluorometer within a 100 meter radius of that station (excluding outliers) was determined. FDA s analyses and conclusions were based upon the lower of these dilution values in a conservative approach. 2.6 Microbiological Analysis of Wastewater Indicator Microorganisms FC densities in the WWTP influent and effluent were determined using a conventional five-tube, three-dilution MPN procedure. MSC densities were determined by using a modified double-agar-overlay method initially described by Cabelli (1988); the E. coli strain HS(pFamp)R (ATCC ) was utilized as the bacterial host strain. Virus concentration and RNA extraction Viral analysis for the sewage utilizes elution with an alkaline buffer followed by ultracentrifugation (Williams-Woods, et al., 2011). Concentrates were extracted for RNA with RNeasy Mini Kit (Qiagen, Valencia, CA) utilizing 6M guanidium isothiocyanate as a lysis solution. Extracted RNA and DNA was tested by real-time reverse transcription (RT)-qPCR and qpcr respectively.

23 RT-qPCR Positive controls used for NoV GI and GII were in vitro RNA transcripts of sequences cloned from positive clinical samples previously identified as NoV (Burkhardt, et al., 2006). Primers and probes for NoV GI and GII targeted the most conserved region of the open reading frame 1 (ORF1)-ORF2 junction. Real-time RT-qPCR for detection of NoV GI and NoV GII with an RNA IAC was performed in a 25-µl reaction volume by using a one-step RT-PCR kit (Qiagen). The primer concentrations for the NoV targets were 300 nm each, and the concentrations for the IAC primers (46F and 194R) were 75 nm each. The 5' nuclease probe concentrations for NoV and the IAC target were 100 and 150 nm each, respectively. The final concentration of MgCl 2 in the real-time RT-qPCR was 4 mm. Thermal cycling was run using the SmartCycler II system with the following conditions: 50 C for 3,000 s and 95 C for 900 s followed by 50 cycles of 95 C for 10 s, 53 C for 25 s, and 62 C for 70 s. Fluorescence was read at the end of the 62 C elongation step. Default analysis parameters were used, except that the manual threshold fluorescence units were set to 10. Samples positive with the initial primer and probe sets for NoV GI and/or NoV GII were subjected to a secondary detection assay. Amplification of the original RNA extract was performed with primers from the B region by conventional RT-PCR (see Table 1 in DePaola, et al., 2010). Amplification of a second region of the genome is non-contiguous to the first and serves as an indication that the RNA was not degraded. Adenovirus The positive control used for Adenovirus (AdV) was serotype 41 isolated from a clinical stool sample, propagated in-house by utilizing the A-549 cell line. Real-time PCR for the detection of AdV was performed in a 25-mL reaction volume by using Platinum TAQ DNA Polymerase (Life Technologies, Grand Island, NY) as previously described with slight modifications (Williams- Woods, et al., 2011). A DNA IAC utilizing the 46F and 194R primers and the TxRed-labeled probe as previously described was added with final primer and probe concentrations of 0.75 mm and 1.5 mm, respectively (DePaola et al., 2010). Cycle parameters were slightly adjusted as follows: 95 o C for 120 s followed by 50 cycles of 95 o C for 3 s, 53 o C for 10 s, and 65 o C for 70 s. AdV primers and probe were previously described with slight modifications to the probe (Heim, 2003) whereby probe was FAM-ZEN labeled as a fluorescent dye on the 5 end and an Iowa Black quencher dye labeled on the 3 end. Fluorescence was read at the end of the 72 C elongation step. Default analysis parameters were used except that the manual threshold fluorescence units were set to 10. Murine norovirus The extraction control used for murine norovirus was purchased from ATCC PTA-5935 and propagated using the RAW264.7 cell line. Real-time RT-qPCR was utilized for the detection of murine norovirus (the extraction control virus) with an RNA IAC in a 25-µl reaction volume by using a one-step RT-PCR kit (Qiagen). Primers and probes were utilized as described in Hewitt, et al., Thermal cycling was run using the SmartCycler II system. Fluorescence was read at the end of the elongation step and the default analysis parameters were used except that the manual threshold fluorescence units were set to 10.

24 Appendix 2

25 Figure 1: Map of Station Locations, Outfalls, and Classified Growing Areas

26 Figure 2: Average Velocity and Travel Time Estimates Based on Dye Data

27 Figure 3: Station 1 - WET Labs 585 Data

28 Figure 4: Station 2 - WET Labs 913 Data

29 Figure 5: Station 4 - WET Labs 2032 Data

30 Figure 6: Station 5 - WET Labs 2416 Data

31 Figure 7: Station 6 - WET Labs 1730 Data

32 Figure 8: Dilution Zones and Accumulated Dye Tracking Results for a 0.75 MGD Flow Based on the March 10 13, 2014 Study Period

33 Figure 9: Dilution Zones and Accumulated Dye Tracking Results Projected for a 1.0 MGD Flow Based on March 10 13, 2014 Study Period

34 Figure 10: Dilution Zones and Accumulated Dye Tracking Results Projected for a 2.0 MGD Flow Based on March 10 13, 2014 Study Period

35 Figure 11: Dilution Zones and Accumulated Dye Tracking Results Projected for a 3.0 MGD Flow Based on March 10 13, 2014 Study Period

36 Table 1: Influent from the Bayou La Batre WWTP FC and MSC Levels Sample Date Time FC (cfu/100 FC Log MSC MSC # Collected ml) (cfu/100 ml) (pfu/100 ml) (MPN/100 ml) 22 3/14/2014 9: not done Table 2: Pre-UV Treated Effluent from the Bayou La Batre WWTP FC and MSC Levels Sample # Date Collected Time FC (cfu/100 ml) FC Log (cfu/100 ml) MSC (pfu/100 ml) 1 3/13/ : /13/ : /13/ : /13/ : /13/ : /13/ : /14/2014 1: /14/2014 3: /14/2014 5: /14/2014 7: /14/2014 9: MSC (MPN/100 ml) Table 3: Post-UV Treated Effluent from the Bayou La Batre WWTP FC and MSC Levels Sample # Date Collected Time FC (cfu/100 ml) FC Log (cfu/100 ml) MSC (pfu/100 ml) 1 3/13/ : /13/ : /13/ : /13/ : /13/ : /13/ : /14/2014 1: /14/2014 3: /14/2014 5: /14/2014 7: /14/2014 9: MSC (MPN/100 ml)

37 Table 4: Influent and Effluent Samples from the Bayou La Batre WWTP Human Virus Levels FDA # Sample # Sample Description GI SC GII SC AdV SC Extraction Efficiency 1 Post 1 Post UV-treated Effluent not detected not detected 8.11E+00 68% 2 Post 3 Post UV-treated Effluent not detected not detected 8.19E+01 not done 3 Post 5 Post UV-treated Effluent not detected not detected 1.57E+02 not done 4 Post 7 Post UV-treated Effluent not detected not detected 1.12E+01 not done 5 Post 9 Post UV-treated Effluent not detected not detected 2.64E+01 not done 6 Post 11 Post UV-treated Effluent not detected not detected not detected not done 7 Post 13 Post UV-treated Effluent not detected not detected not detected not done 8 Post 15 Post UV-treated Effluent not detected not detected not detected not done 9 Post 17 Post UV-treated Effluent not detected not detected not detected not done 10 Post 19 Post UV-treated Effluent not detected not detected 1.72E+02 not done 11 Post 21 Post UV-treated Effluent not detected not detected not detected not done 12 Blank 1 DI Water not detected not detected not detected not done 13 Pre 1 Pre UV-treated Effluent not detected not detected 9.39E+01 not done 14 Pre 3 Pre UV-treated Effluent not detected not detected not detected not done 15 Pre 5 Pre UV-treated Effluent not detected not detected not detected 69% 16 Pre 7 Pre UV-treated Effluent not detected not detected not detected not done 17 Pre 9 Pre UV-treated Effluent not detected not detected 1.52E+03 79% 18 Pre 11 Pre UV-treated Effluent not detected not detected 1.48E+01 not done 19 Pre 13 Pre UV-treated Effluent not detected not detected not detected not done 20 Pre 15 Pre UV-treated Effluent not detected not detected not detected 75% 21 Pre 17 Pre UV-treated Effluent not detected not detected not detected not done 22 Pre 19 Pre UV-treated Effluent not detected not detected 8.02E+01 not done 23 Pre 21 Pre UV-treated Effluent not detected not detected not detected not done 24 Blank 2 DI Water not detected not detected not detected not done 25 Inf Influent 1.17E+02 not detected 1.38E+03 60% not detected not detected not detected not done 27 Blank 3 DI Water not detected not detected not detected 71% *Extraction Efficiency: Log(spike)/Log(recovered) x efficiency is not used to adjust virus recovery. ** All numbers are reported in RTqPCR units/100ml of sewage. *** PCR controls were not added into the table. All NTC were negative, Positives were positive, and IAC Showed no evidence of inhibition. **** Standard deviation of IAC throughout all samples is 1.06.

38 Appendix 3

39 Figure 1: Dye Tracking Results for a 0.75 MGD Flow Based on March 10 13, 2014 Study Period

40 Figure 2: Dye Tracking Results Projected for a 1.0 MGD Flow Based on March 10 13, 2014 Study Period

41 Figure 3: Dye Tracking Results Projected for a 2.0 MGD Flow Based on March 10 13, 2014 Study Period

42 Figure 4: Dye Tracking Results Projected for a 3.0 MGD Flow Based on March 10 13, 2014 Study Period

43 Figure 5: Distance vs Dilution East of Outfall (minimum dilutions shown)

44 Figure 6: Distance vs Dilution West of Outfall (minimum dilutions shown)

45 Figure 7: Dilution vs Distance Regression East of Outfall

46 Figure 8: Dilution vs Distance Regression West of Outfall

47 Figure 9: 100 Meter Buffer Frequency of Outfall