IN-CANAL PHOSPHORUS TREATMENT STUDY FOR BARR LAKE

Transcription

1 IN-CANAL PHOSPHORUS TREATMENT STUDY FOR BARR LAKE Final Report: Conceptual Designs and Costs for Evaluated Treatment Options December 2014 Prepared For: Barr Lake/Milton Reservoir Watershed Association 6450 York Street Denver, CO Prepared By: Environmental Research and Design, Inc Trentwood Blvd., Suite 102 Belle Isle (Orlando), FL Phone: Harvey H. Harper, III, Ph.D., P.E.-Project Director

2 TABLE OF CONTENTS Section LIST OF FIGURES LIST OF TABLES Page LF-1 LT-1 1. INTRODUCTION EXISTING CHARACTERISTICS OF BURLINGTON DITCH INFLOWS TO BARR LAKE Discharge Characteristics Water Quality Characteristics General Parameters Nutrients Mass Loadings FEASIBILITY EVALUATION FOR ALUM TREATMENT OF BARR LAKE INFLOWS Sample Collection Laboratory Activities Jar Testing Laboratory Methods Jar Test Results Floc Generation and Accumulation Effectiveness and Impacts of Alum Treatment EVALUATION OF POTENTIAL TREATMENT OPTIONS Load Reduction Goals Evaluation Basis Proposed Treatment System Loading Alum Treatment of Ditch Inflows Special Requirements Basic Alum Treatment System Components Design Parameter Summary Conceptual Cost Estimates for Basic System Components Floc Removal Options Conceptual Floc Collection Designs Annual O&M Costs Phosphorus Removal Costs Treatment System Operation 4-37 TOC-1

3 TABLE OF CONTENTS -- CONTINUED Section Page 4.5 Media Filtration Basic Media Filtration System Components Conceptual Design and Costs Annual O&M Costs Present Worth and Phosphorus Removal Costs Treatment System Operation SUMMARY 5-1 Appendices A. Results of Laboratory Jar Testing Conducted on Burlington Ditch Inflows to Barr Lake B. Section 2 of the Technical Report Titled Technical Assistance for the Northern Everglades Chemical Treatment Pilot Project (July 6, 2009) TOC-2

4 LIST OF FIGURES Figure Number / Title Page 1-1 Location Map for Barr Lake and Ditch/Canal System Aerial Photograph of Barr Lake Recorded Discharges to Barr Lake Through the Burlington Ditch for Water Yeas Frequency Distribution of Inflows to Barr Lake Cumulative Frequency Distribution of Inflows to Barr Lake Measured ph Values in the Burlington Ditch Inflow to Barr Lake from Measured Alkalinity Values in the Burlington Ditch Inflow to Barr Lake from Measured Conductivity Values in the Burlington Ditch Inflow to Barr Lake from Measured Sulfate Values in the Burlington Ditch Inflow to Barr Lake from Seasonal Variability in Inflow Concentrations of ph, Alkalinity, Conductivity, and Temperature Measured Values of Total Nitrogen Species in the Burlington Ditch Inflow to Barr Lake from Measured Values of Total Phosphorus Species in the Burlington Ditch Inflow to Barr Lake from Seasonal Variability in Inflow Concentrations of Nutrients Calculated Mass Loadings of Total Nitrogen from the Burlington Ditch to Barr Lake from Calculated Mass Loadings of Total Phosphorus from the Burlington Ditch to Barr Lake from Laboratory Jar Test Apparatus Settling Cones with Accumulated Alum Floc from Different Alum Doses 3-5 LF-1

5 LIST OF FIGURES -- CONTINUED Figure Number / Title Page 3-3 Graphical Relationship Between Floc Generation and Alum Dose Alum Floc Drying Process Daily Phosphorus Loadings to Barr Lake from Daily Phosphorus Loadings to Barr Lake to Achieve 75% Annual Total Phosphorus Reduction Parcel Ownership Along the Burlington Ditch and the Bifurcation Structure Conceptual Design of an Off-Line Channel for Alum Addition Conceptual Schematic of Alum Treatment System Components for the 2,500 kg TP/yr Removal Option Conceptual Schematic of Alum Treatment System Components for the 14,500 kg TP/yr Removal Option Conceptual Schematic of Alum Treatment System Components for the 53,250 kg TP/yr and 68,160 kg TP/yr Removal Options Typical Settling Pond Design for the Barr Lake Inflow Floc Discharging into Barr Lake During Low Water Conditions Floc Discharging into Barr Lake During High Water Conditions Assumed Annual Floc Depositional Areas in Barr Lake Schematic of Floc Removal System for Full Floc Removal with the 2,500 kg TP/yr Treatment Option Schematic of Floc Removal System for Full Floc Removal with the 14,500 kg TP/yr Treatment Option Cross-Section Through Media Filtration System Cross-Section of Media Portion of Filter System Conceptual Schematic of an Off-Line Media Filtration System for the 2,500 kg TP/yr Treatment Option Conceptual Schematic of an Off-Line Media Filtration System for the 14,500 kg TP/yr Treatment Option 4-47 LF-2

6 LIST OF TABLES Table Number / Title Page 2-1 Summary of Recorded Inflows to Barr Lake by Discharge Rate Analytical Methods and Detection Limits for Laboratory Analyses Conducted on Jar Test Samples Mean Phosphorus Removal Efficiencies in Barr Lake Inflow Jar Tests Mean Water Quality Characteristics of Barr Lake Inflow Jar Tests Summary of Floc Generation Data Summary of Evaluated Load Reduction Scenarios for Barr Lake Estimated Process Removal Efficiencies for the Alum Treatment and Media Filtration Options Summary of Treatment Design Criteria Current Parcel Listings in the Vicinity of Barr Lake (September 2014) Conceptual Cost Estimate for the Off-Line Channel Option Alum Inflow Treatment Design Parameters for the Evaluated Load Reduction Options Conceptual Cost Estimates for Basic Alum Treatment Components for the Evaluated Load Reduction Options Pond Design Parameters for the Full Floc Collection Options Pond Design Parameters for the Partial Floc Collection Options Summary of In-Lake Floc Accumulation in Barr Lake for the Full and Partial Floc Discharge Options Conceptual Construction Costs for Floc Collection with the 2,500 kg TP/yr Removal Option Conceptual Construction Costs for Floc Collection with the 14,500 kg TP/yr Removal Option 4-28 LT-1

7 LIST OF TABLES -- CONTINUED Table Number / Title Page 4-13 Conceptual Construction Costs for Floc Collection with the 53,250 kg TP/yr Removal Option Conceptual Construction Costs for Floc Collection with the 68,160 kg TP/yr Removal Option Estimated Annual O&M Costs for the Alum Inflow Treatment Options Calculated Phosphorus Removal Costs for the 2,500 kg TP/yr Removal Option Calculated Phosphorus Removal Costs for the 14,500 kg TP/yr Removal Option Calculated Phosphorus Removal Costs for the 53,250 kg TP/yr Removal Option Calculated Phosphorus Removal Costs for the 68,160 kg TP/yr Removal Option Operational Procedures for the Alum Inflow Treatment Option Media Filtration System Design Parameters for the Evaluated Load Reduction Options Conceptual Cost Estimate for Media Filtration with the 2,500 kg TP/yr Removal Option Conceptual Cost Estimate for Media Filtration with the 14,500 kg TP/yr Removal Option Conceptual Cost Estimate for Media Filtration with the 53,250 kg TP/yr Removal Option Estimated Annual O&M Costs for the Media Filtration Options Calculated Present Worth and Phosphorus Removal Costs for the Media Filtration Option Operational Procedures for the Media Filtration Option Summary of Construction, O&M, and Total Phosphorus Mass Removal Costs for the Alum Inflow Treatment Options Summary of Construction, O&M, and Total Phosphorus Mass Removal Costs for the Media Filtration Option 5-2 LT-2

8 SECTION 1 INTRODUCTION This report provides a summary of work efforts performed by Environmental Research & Design, Inc. (ERD) for the Barr Lake/Milton Reservoir Watershed Association (BMW) to evaluate potential treatment strategies to reduce phosphorus loadings delivered to Barr Lake through the Burlington Ditch. A location map for Barr Lake and the connecting canal system is given on Figure 1-1. Barr Lake is located approximately 14 miles northeast of downtown Denver, north of the Denver International Airport. Barr Lake is a 2,000-acre man-made reservoir which is used for irrigation and municipal water supplies, although the lake also provides wildlife habitat and recreational opportunities. An aerial photograph of Barr Lake is given on Figure 1-2. Barr Lake currently receives substantially elevated mass loadings of both total nitrogen and total phosphorus which have led to accelerated algal growth within the lake and elevated ph levels. Barr Lake was listed on the (d) list as an impaired waterbody due to exceedances of the upper ph standard of 9.0 units. According to AMEC (2008), nutrient loadings to Barr Lake currently average approximately 71,000 kg/yr of total phosphorus and 600,000 kg/yr of total nitrogen. The Barr Lake/Milton Reservoir Watershed Association contracted with ERD during January 2014 to conduct a feasibility evaluation of potential options for reducing phosphorus loadings discharging into Barr Lake through the Burlington Ditch. The stated objective of this project is to identify and evaluate both passive and active treatment options capable of achieving specified phosphorus load reductions, including conceptual designs and cost estimates. The options must consider the timing of the inflows to Barr Lake, which occur primarily during the late-fall and winter months from November-March, and minimize water losses from evapotranspiration. After identifying potential options, ERD conducted an initial evaluation and ranking of the proposed options to assist in developing a short list of the top ranked treatment options. An Interim Report, prepared and submitted by ERD during July 2014, outlined the evaluated BMPs and provides the results of the ranking procedure. This analysis identified alum treatment of the Burlington Ditch inflow and filtration using biologically activated media (media) as potential treatment methods for reducing phosphorus loadings to the lake. The results of the rankings were presented to the BMW Technical Committee which concurred with the findings and authorized ERD to continue further evaluations for the two treatment options. Detailed evaluations were conducted for the alum treatment and media treatment options which included a conceptual site design, estimates of land requirements, construction and O&M costs, and anticipated annual phosphorus removal and mass removal costs. The results of this detailed analysis form the basis of this current document. 1-1

9 1-2 Barr Lake Burlington Ditch Inflow Figure 1-1. Location Map for Barr Lake and Ditch/Canal System.

10 1-3 Figure 1-2. Aerial Photograph of Barr Lake.

11 SECTION 2 EXISTING CHARACTERISTICS OF BURLINGTON DITCH INFLOWS TO BARR LAKE As outlined in the stated objectives for the project, treatment options designed to produce in-canal phosphorus reductions for the Burlington Ditch must consider the site-specific hydrologic and water quality characteristics of the Burlington Ditch inflows and the climatic and meteorological characteristics of the region in general. A discussion of existing characteristics of the Burlington Ditch is given in the following sections. 2.1 Discharge Characteristics Information on recorded discharge rates into Barr Lake from the Burlington Ditch were provided to ERD by BMW for water years 2006, 2007, 2008, 2009, 2010, 2011, and A graphical summary of recorded average daily discharges to Barr Lake is given on Figure 2-1. In general, inflows into Barr Lake are highly variable, ranging from 0 cfs to almost 500 cfs over the available period of record, although the vast majority of recorded inflows appear to be approximately 200 cfs or less. The available data for the seven water years summarized on Figure 2-1 cover a total of 2,556 days. A summary of recorded inflows to Barr Lake by discharge rate during the 2,556-day period is given in Table 2-1. Discharge rates into Barr Lake were essentially 0 cfs during 684 of the 2,556 days, representing approximately 26.8% of the period. A relatively small number of events are characterized by flow rates in the range of cfs, cfs, and 5-10 cfs, which combined together reflect only 8.1% of the period. Inflow rates ranging from cfs and cfs occur approximately 11.0% and 11.3% of the time, respectively. The highest frequency of inflows to Barr Lake, other than the 0 cfs discharge category, occur for the cfs and cfs discharge rates which occur 17.1% and 20.0% of the time, respectively. Approximately 74.2% of the inflows are less than 100 cfs, with 94.2% less than 200 cfs. Inflow rates into Barr Lake in excess of 200 cfs occur relatively infrequently, reflecting approximately 5.8% of the data. A graphical frequency distribution of inflows to Barr Lake is given on Figure 2-2. A cumulative frequency distribution of inflows to Barr Lake is given on Figure 2-3 based upon the information summarized in Table 2-1. The cumulative frequency distribution is useful in evaluating the anticipated performance efficiency for treatment options by providing relationships between the frequency of occurrence and discharge rates. This information is used in subsequent sections to evaluate the selected BMP treatment options. 2-1

12 Inflow (cfs) Year Figure 2-1. Recorded Discharges to Barr Lake Through the Burlington Ditch for Water Years INFLOW INTERVAL (cfs) TABLE 2-1 SUMMARY OF RECORDED INFLOWS TO BARR LAKE BY DISCHARGE RATE NUMBER OF DAYS CUMULATIVE NUMBER OF DAYS PERCENT OF EVENTS CUMULATIVE PERCENT (%) , , , , , , > ,

13 2-3 Figure 2-2. Frequency Distribution of Inflows to Barr Lake. Figure 2-3. Cumulative Frequency Distribution of Inflows to Barr Lake.

14 Water Quality Characteristics General Parameters Information on water quality characteristics for discharges through the Burlington Ditch into Barr Lake were provided to ERD by BMW for the period from A graphical summary of measured ph values in the Burlington Ditch inflow to Barr Lake is given on Figure 2-4. Measured ph values in the inflow are highly variable, ranging from approximately over the available period of record. The vast majority of measured ph values appear to be within the range of approximately Any proposed treatment process would need to be operational under a relatively wide range of ph values ph Year Figure 2-4. Measured ph Values in the Burlington Ditch Inflow to Barr Lake from A graphical summary of measured alkalinity values in the Burlington Ditch inflow to Barr Lake from is given on Figure 2-5. Measured alkalinity values in the inflow are also highly variable, ranging from approximately 50 mg/l to more than 300 mg/l, although the vast majority of measured values appear to range from approximately mg/l, reflecting moderately to well buffered conditions. Levels of alkalinity within the water column will be a significant factor in evaluating the potential use of alum for reducing phosphorus loadings in the inflow since alum addition results in consumption of alkalinity as part of the reaction process. The historical alkalinity values illustrated on Figure 2-5 are more than adequate to support an alum treatment process, especially during the period from November-March when the majority of inflows occur.

15 Alkalinity (mg/l) Year Figure 2-5. Measured Alkalinity Values in the Burlington Ditch Inflow to Barr Lake from A graphical summary of measured conductivity values in the Burlington Ditch inflow to Barr Lake from is given on Figure 2-6. Measured conductivity values have been highly variable, ranging from approximately 450-2,200 mho/cm, although the vast majority of measured values appear to be between 500-1,100 mho/cm. Conductivity values in this range are typical of canal drainage systems receiving significant wastewater inputs. A graphical summary of measured sulfate values in the Burlington Ditch inflow to Barr Lake from is given on Figure 2-7. Measured sulfate concentrations in the ditch have ranged from approximately 40 mg/l to more than 500 mg/l, although the vast majority of measured values appear to be between mg/l. Sulfate concentrations do not pose an issue with the selected treatment BMP processes, although elevated sulfate concentrations may pose a risk in anoxic portions of the media filter system for increasing rates of production of methyl mercury due to sulfate-reducing bacteria.

16 Conductiviy (µmho/cm) Year Figure 2-6. Measured Conductivity Values in the Burlington Ditch Inflow to Barr Lake from Sulfate (mg/l) Year Figure 2-7. Measured Sulfate Values in the Burlington Ditch Inflow to Barr Lake from

17 2-7 A graphical summary of seasonal variability in inflow concentrations of ph, alkalinity, conductivity, and temperature to Barr Lake is given on Figure 2-8. Measured ph values in inflows to Barr Lake appear to be lowest during January, with an overall mean ph for the month of approximately 7.8. However, measured inflow ph values for the remaining months typically exceed 8.0, with more elevated ph values observed during spring and fall conditions and lower ph values observed during summer conditions. A similar pattern is also apparent for alkalinity which exhibits mean monthly concentrations of approximately 150 mg/l or more during fall, winter, and early-spring, with somewhat lower concentrations observed during late-spring and summer conditions. A similar pattern is also apparent for conductivity, with the most elevated conductivity values observed during late-fall, winter, and early-spring and substantially lower values observed during summer conditions. An inverse relationship is observed for temperature, with the lowest mean monthly temperatures, typically approximately 5-6 o C or less, observed during winter conditions and higher temperatures observed during remaining portions of the year Nutrients A graphical summary of measured concentrations of nitrogen species in the Burlington Ditch inflow to Barr Lake is given on Figure 2-9. Measured concentrations of total nitrogen in the Burlington Ditch inflow have been highly variable over the available period of record from , ranging from approximately 2-16 mg/l, although the vast majority of total nitrogen concentrations appear to be within the range of approximately 2-5 mg/l. Total nitrogen concentrations appear to exhibit a seasonal effect, with the most elevated levels of total nitrogen occurring during late-fall and winter conditions. NO x (nitrite + nitrate) represents a significant portion of the total nitrogen present during much of the year, although TKN concentrations exceed NO x concentrations at times. A graphical summary of measured phosphorus species in the Burlington Ditch inflow to Barr Lake over the period from is given on Figure Measured concentrations of total phosphorus have also been highly variable, with values ranging from mg/l. The vast majority of the total phosphorus is present as soluble reactive phosphorus (SRP), with very little contribution from organic or particulate phosphorus forms. A distinct seasonal pattern is apparent for concentrations of total phosphorus, with the most elevated concentrations observed during late-fall and winter conditions. Since this is the period during which water enters Barr Lake, the treatment systems must be capable of treating relatively elevated levels of total phosphorus which are primarily present in a soluble inorganic form. A graphical summary of seasonal variability in inflow concentrations of nutrients to Barr Lake is given on Figure In general, monthly concentrations of nitrate, total nitrogen, SRP, and total phosphorus appear to follow a similar pattern, with more elevated concentrations for each parameter observed during fall, winter, and early-spring conditions, and substantially lower concentrations during remaining portions of the year. Nutrient concentrations appear to be greatest during periods of the year when Barr Lake is typically receiving inflow, reflecting typical concentrations which would be treated by the proposed inflow treatment system.

18 ph 250 Alkalinity ph Month Month Figure 2-8. Seasonal Variability in Inflow Concentrations of ph, Alkalinity, Conductivity, and Temperature. Alkalinity ph (s.u.) Alkalinity (mg/l) 1,400 1,200 1, Conductivity Temp Cond Month Month Temp Spec. Cond. (µmho/cm) Temp. ( C)

19 NO x TKN Total N Nitrogen (mg/l) Year Figure 2-9. Measured Values of Total Nitrogen Species in the Burlington Ditch Inflow to Barr Lake from SRP Total P Phosphorus (mg/l) Year Figure Measured Values of Total Phosphorus Species in the Burlington Ditch Inflow to Barr Lake from

20 SRP Total P SRP Month Month Figure Seasonal Variability in Inflow Concentrations of Nutrients. Total P SRP (mg/l) Nitrate (mg/l) Nitrate Total N Nitrate Month Month Total N Nitrate (mg/l) Total N (mg/l)

21 Mass Loadings A graphical summary of calculated mass loadings of total nitrogen from the Burlington Ditch to Barr Lake from is given on Figure The information provided in this figure was obtained by multiplying daily measured flow rates times measured water column concentrations of total nitrogen for each day on which simultaneous measurements of discharge and water quality are available. Mass loadings of nitrogen are highly variable as a result of the variability in inflow rates and total nitrogen concentrations. However, nitrogen loadings are clearly greatest during late-fall and winter conditions, with loadings during this period several times greater than loadings during other portions of the year Total N (kg/day) Year Figure Calculated Mass Loadings of Total Nitrogen from the Burlington Ditch to Barr Lake from A graphical summary of calculated mass loadings of total phosphorus from the Burlington Ditch to Barr Lake from is given on Figure The values presented in Figure 2-13 were calculated by multiplying measured daily discharge rates times measured concentrations of total phosphorus for each day which contained simultaneous discharge and water quality measurements. A distinct seasonal pattern is apparent for mass loadings of total phosphorus, with more elevated loadings observed during late-fall and winter conditions compared with other portions of the year. Any proposed BMP treatment process must be capable of handling both the variability and magnitude of mass loadings for total phosphorus entering Barr Lake.

22 Total P (kg/day) Year Figure Calculated Mass Loadings of Total Phosphorus from the Burlington Ditch to Barr Lake from

23 SECTION 3 FEASIBILITY EVALUATION FOR ALUM TREATMENT OF BARR LAKE INFLOWS The scope of services for this project specifically requested that alum treatment of inflows into Barr Lake be included as a potential option for reducing nutrient loadings to the lake. The feasibility of using alum for treatment of inflows is highly dependent upon the specific water quality characteristics of the inflow water which dictates the anticipated removal efficiencies, floc settling rates, and floc generation rates. Therefore, laboratory jar testing was conducted on samples collected from the Burlington Ditch inflow to Barr Lake to provide sitespecific information on the potential feasibility. 3.1 Sample Collection Grab samples of inflows from the Burlington Ditch into Barr Lake were collected by Mr. Steve Lundt on three separate occasions for laboratory jar testing. Separate grab samples were collected on January 28, March 14, and April 15, 2014, with a sample volume of approximately 5 gallons collected each event. The grab samples were packed in ice and shipped to the ERD Laboratory in Orlando, FL for subsequent testing. The three samples are intended to represent the potential range of water quality characteristics of Barr Lake inflows during a typical inflow cycle Jar Testing 3.2 Laboratory Activities Laboratory jar tests were conducted using alum on each of the composite Burlington Ditch inflow water samples collected during the field monitoring program. The objective of the jar tests is to evaluate the water quality response of alum coagulation at various doses on ditch inflows collected at each site. The laboratory jar tests were conducted using a Phipps and Bird jar test apparatus. A photograph of the jar test apparatus is given on Figure 3-1. Jar tests were conducted on each sample at six separate alum doses, including 2, 4, 6, 8, 10, and 12 mg Al/liter, to evaluate a wide range of potential application doses. 3-1

24 3-2 Figure 3-1. Laboratory Jar Test Apparatus. Each of the jar test beakers was filled with 2 liters of tributary water from a particular collection date. The paddle wheels were then activated and allowed to rotate at a constant speed of 60 rpm. The desired alum dose was then added to each container, and mixing was continued for a period of 60 seconds. At the end of the mixing period, the paddle wheels were stopped, removed from the beakers, and the resulting mixture allowed to settle. Measurements of ph were conducted initially in the raw sample and in each of the alum treated samples approximately 1 minute after addition of the selected alum dose and again after a period of 1 hour which generally reflects the minimum ph value achieved for a specific alum dose. Additional measurements of ph were recorded 24 hours after addition of the coagulant to document changes in ph which typically occur after alum addition. The alum treated samples were then allowed to settle for a period of 24 hours, simulating settling processes which would occur within the water column of a settling pond or lake. During the settling process, visual observations were recorded of floc settling rates for each alum dose, along with the time required for complete settling of the alum floc if it occurred prior to the end of the 24-hour period. This information is useful in designing and sizing settling ponds for floc removal. At the end of the 24-hour settling period, the clear supernatant was decanted from each jar test container for subsequent laboratory analyses Laboratory Methods Each of the samples generated during the laboratory jar test procedures, including both raw and treated samples at each coagulant dose, was analyzed for a wide variety of chemical constituents, including general parameters, nutrients, and aluminum. A summary of analytical methods and detection limits for analyses conducted by ERD on each of the generated jar test samples is given in Table 3-1. The ERD Laboratory is NELAC-certified (No. E ) for each of the parameters listed in Table 3-1.

25 3-3 TABLE 3-1 ANALYTICAL METHODS AND DETECTION LIMITS FOR LABORATORY ANALYSES CONDUCTED ON JAR TEST SAMPLES General Parameters Nutrients MEASUREMENT PARAMETER Hydrogen Ion (ph) Specific Conductivity Alkalinity Color Turbidity Ammonia-N (NH 3 -N) Nitrate + Nitrite (NO x -N) Total Nitrogen Orthophosphorus (SRP) Total Phosphorus METHOD SM-21 2, Sec H + B EPA-83 3, Sec SM-21, Sec B SM-21, Sec C SM-21, Sec B SM-21, Sec NH 3 G SM-21, Sec NO 3 F SM-21, Sec N C SM-21, Sec P F SM-21, Sec P B5 METHOD DETECTION LIMITS (MDLs) 1 N/A 0.1 mho/cm 0.5 mg/l 1 Pt-Co unit 0.3 NTU 5 g/l 5 g/l 10 g/l 1 g/l 1 g/l Metals Diss. Aluminum SM-21, Sec Al E 0.8 g/l NOTES: 1. MDLs are calculated based on the EPA method of determining detection limits 2. Standard Methods for the Examination of Water and Wastewater, 21 st Edition, Methods for Chemical Analysis of Water and Wastes, EPA 600/ , Revised March Jar Test Results A complete listing of the results of laboratory analyses conducted on jar test samples collected from Barr Lake inflows on January 28, March 14, and April 15, 2014 is provided in Appendix A. A tabular summary of mean phosphorus removal efficiencies achieved in the Barr Lake inflow jar test samples is given in Table 3-2. Alum treatment of Burlington Ditch inflows to Barr Lake was extremely effective in removing both SRP and total phosphorus from the test samples. Removal efficiencies ranging from 75-87% were achieved for SRP, with removal efficiencies ranging from 74-82% for total phosphorus even at the lowest tested alum dose of 2 mg Al/liter. However, alum treatment at this dose resulted in a precipitate which exhibited relatively slow settling characteristics. Small additional improvements in phosphorus removal efficiencies were obtained with increasingly higher alum doses.

26 3-4 TABLE 3-2 MEAN PHOSPHORUS REMOVAL EFFICIENCIES IN BARR LAKE INFLOW JAR TESTS SAMPLE DESCRIPTION TEST NO. 1 TEST NO. 2 TEST NO. 3 SRP Total P SRP Total P SRP Total P Raw mg Al/liter Al/liter Al/liter Al/liter > Al/liter > Al/liter > > A summary of mean water quality characteristics of alum treated Barr Lake inflow samples is given on Table 3-3. The addition of alum to Barr Lake inflow samples had relatively minimal impact on measured concentrations of nitrogen species. However, alum addition resulted in significant reductions in measured concentrations of phosphorus species, with SRP concentrations reduced from a mean raw concentration of 352 g/l to a mean treated concentration of only 5 g/l at an alum dose of 6 mg Al/liter with little additional improvement at higher doses. Similarly, particulate phosphorus concentrations were reduced from a mean raw concentration of 75 g/l to 17 g/l at an alum dose of 6 mg Al/liter. A significant reduction was also observed in measured concentrations of total phosphorus, which was reduced from a mean of 452 g/l in the raw samples to only 26 g/l in the alum treated samples at a dose of 6 mg Al/liter with relatively minimal reductions at higher doses. Based upon the water quality characteristics summarized in Table 3-3, an optimum alum dose of approximately 6 mg Al/liter appears to be appropriate for treatment of Barr Lake inflows. Water quality improvements obtained at alum doses in excess of 6 mg Al/l were only marginally lower in value and do not appear to be worth the additional chemical costs to achieve a minimal improvement in removal efficiencies. Alum addition to Barr Lake inflows at the recommended optimum dose of 6 mg Al/liter resulted in an equilibrium ph value of approximately 6.96, with a mean alkalinity of approximately 112 mg/l. It is apparent that alum treatment of Barr Lake inflows will provide substantial reductions in measured concentrations of phosphorus species without undesirable reductions in ph and alkalinity.

27 3-5 TABLE 3-3 MEAN WATER QUALITY CHARACTERISTICS OF BARR LAKE INFLOW JAR TESTS PARAMETER SAMPLE DESCRIPTION Initial ph 1 minute Alkalinity (mg/l) Conductivit y ( mho/cm) NH3 ( g/l) NOx ( g/l) Particulate N ( g/l) Total N ( g/l) SRP ( g/l) Particulate P ( g/l) Total P ( g/l) Color (Pt-Co) Raw , , , mg Al/liter , , , Al/liter , , , Al/liter , , , Al/liter , , , Al/liter , , , Al/liter , , , Floc Generation and Accumulation A significant part of the jar test procedure for evaluating the feasibility of alum treatment is to evaluate floc generation and accumulation rates. Floc generated from each of the three jar tests were combined together by alum dose to provide estimates of the total floc generated as a result of alum treatment for each of the six evaluated aluminum doses. The floc generated from alum treatment at each dose was placed into a settling cone, and the floc was allowed to consolidate for a period of approximately 30 days following addition of the floc generated during the final laboratory jar test. Photographs of the settling cones with the accumulated alum floc from different alum doses are given in Figure 3-2. Figure 3-2. Settling Cones with Accumulated Alum Floc from Different Alum Doses.

28 3-6 A summary of the floc generation data resulting from the three jar test samples is given in Table 3-4. The total volume of treated water for each dose is 6 liters which results from treatment of 2 liters from each of the three separate grab samples. The final collected floc volume is also provided, along with the calculated floc generation rate as a percentage of the treated flow. The values summarized in the final column of Table 3-4 reflect a floc settling time of approximately 30 days. However, over time, the floc will continue to consolidate, with an additional 25% reduction in volume anticipated after a period of several months. ALUM DOSE (mg Al/Liter) TABLE 3-4 SUMMARY OF FLOC GENERATION DATA TREATED VOLUME (liter) FLOC VOLUME (ml) GENERATION RATE (% of Treated Flow) NOTE: Floc exhibits an additional 25% reduction over time A graphical summary of the relationship between floc generation and alum dose is given on Figure 3-3. The relationship appears to be relatively linear, with increasing floc generated as the applied alum dose increases. The information summarized in Figure 3-3 and in Table 3-4 is used in subsequent sections to evaluate floc generation and removal options for the evaluated treatment systems. Figure 3-3. Graphical Relationship Between Floc Generation and Alum Dose. Floc Gen. (% of Treated Flow) Alum Dose (mg Al/L)

29 3-7 The floc generation characteristics illustrated on Figure 3-3 apply only to floc maintained in a wet environment. Settled alum floc contains a moisture content of at least 95%. However, when alum floc is dried, a substantial consolidation occurs with the remaining residual solids representing less than 5% of the original wet floc volume. Photographs of a typical alum floc drying process are given on Figure 3-4. After initial water decanting, the floc cake begins to crack and separate similar to a wastewater sludge. Complete drying of the floc requires approximately 30 days, with the final residue consisting of a hard material that is extremely resistant to dewatering and dissolution. The process illustrated on Figure 3-4 represents a typical sequence of events which would occur when alum floc is applied to a drying bed for consolidation. Floc color is a function of the materials removed from the treated water Floc after initial water decanting (~ 95% moisture) After 4-7 days After 30 days Once completely dried, the floc forms into a rock hard material that will not re-dissolve Figure 3-4. Alum Floc Drying Process.

30 Effectiveness and Impacts of Alum Treatment The concept of using alum to treat stormwater inflows was originated during 1982 by Dr. Harvey H. Harper, P.E. as part of a U.S. EPA-funded Clean Lakes Project on a small urban lake in Orlando, FL. The first practical application to use alum for treatment of stormwater inflows occurred at Lake Ella in Tallahassee, FL. Since that time, more than 60 alum inflow treatment systems have been constructed within the State of Florida, with more than 95% of these projects designed and implemented by ERD. Over the years, ERD has amassed a substantial amount of information related to removal effectiveness, sediment impacts, benthic impacts, and operation and maintenance characteristics for these systems. This information was recently summarized in a report prepared for the South Florida Water Management District as part of an evaluation of the potential for using alum for improvement of waters discharging into the Northern Everglades. This document, titled Technical Assistance for the Northern Everglades Chemical Treatment Pilot Project, was issued during May The section specifically related to alum treatment (Section 2) was written and prepared by ERD and provides a detailed evaluation of the current knowledge related to the use of alum for treatment of runoff inflows. This section is attached to this report as Appendix B and includes an overview of common coagulants, process chemistry, aluminum solubility, impacts from redox potential and ph, applications and success stories, water quality improvements, floc accumulation, floc stability, sediment chemistry, construction costs, and pollutant removal costs.

31 SECTION 4 EVALUATION OF POTENTIAL TREATMENT OPTIONS 4.1 Load Reduction Goals As requested by BMW, load reduction options were evaluated for the alum treatment and media filtration options to reduce phosphorus loadings to Barr Lake based upon specified load reduction scenarios. A summary of the evaluated load reduction scenarios for Barr Lake is given in Table 4-1. Four separate options were evaluated for both alum treatment and media filtration. The first option (Option 1), which specifies a required annual total phosphorus reduction of 2,500 kg/yr, is based upon a 75% reduction of loadings from Cherry Creek, Bear Creek, and Chatfield Reservoirs which have a combined total phosphorus input of approximately 3,000 kg/yr. Option 2 includes a 75% reduction of loadings from Cherry Creek, Bear Creek, and Chatfield Reservoirs plus point source loadings discharging to Barr Lake. This option has a required annual total phosphorus load reduction of approximately 14,500 kg/yr. Option 3 provides for a 75% removal of the total annual phosphorus loading to Barr Lake from the Burlington Canal, equivalent to an annual total phosphorus reduction of 53,250 kg/yr. The final option (Option 4) provides for a removal of 96% of the total annual phosphorus loadings to Bear Lake which is equivalent to the total TMDL required load reductions to the lake. The annual total phosphorus load reduction for this option is 68,160 kg/yr. Each of these four load reduction scenarios were evaluated for both the alum treatment and media filtration options. OPTION NO. 1 2 TABLE 4-1 SUMMARY OF EVALUATED LOAD REDUCTION SCENARIOS FOR BARR LAKE OPTION DESCRIPTION 75% reduction of loadings from Cherry Creek, Bear Creek, and Chatfield Reservoirs (~3,000 kg/yr) 75% reduction of loadings from Cherry Creek, Bear Creek, and Chatfield Reservoirs plus point source loadings REQUIRED ANNUAL TOTAL PHOSPHORUS REDUCTION (kg/yr) 2,500 14, % removal of total annual load 53, % removal of total annual load (TMDL required reduction) ,160

32 Evaluation Basis Due to the large variability in phosphorus concentrations in the inflows to Barr Lake, mass load reductions to the lake cannot be calculated based simply upon a treatment of corresponding fraction of hydrologic inputs but must be based on a simulation involving both flow and phosphorus concentrations. A graphical summary of variability in phosphorus loadings to Barr Lake over the period from is illustrated on Figure 4-1. The information indicated on this figure was obtained by multiplying total phosphorus concentrations times the measured flow rate on days for which phosphorus concentration data are available. Daily phosphorus loadings to Barr Lake range from zero to more than 900 kg/day over the available period of record. A simulation model was developed by ERD to evaluate the required inflow discharge rate which must be diverted for treatment to achieve the specified load reduction goals summarized in Table 4-1. A summary of assumed process removal efficiencies for the alum treatment and media filtration options is given in Table 4-2. An efficiency of 90% is assumed for the alum treatment option which appears to be a conservative estimate since the jar test results indicate removal efficiencies ranging from 92-95% for total phosphorus at the recommended optimum dose of 6 mg Al/liter. An assumed phosphorus removal efficiency of 80% is assumed for media filtration based upon typical literature values. The simulation model assumes that the listed removal efficiencies in Table 4-2 are applied to water which is diverted from the Burlington Canal for treatment. The model then provides estimates of load reductions on a daily basis over the available period of record from and provides a summary of the estimated mass of phosphorus removed from the inflow over the 7-year simulation period. The rate at which water is diverted into the treatment system can then be varied until the desired annual load reduction is achieved. A summary of the results of the simulation model for estimating inflow diversion volumes required to achieve a 75% overall load reduction for inputs to Barr Lake, assuming a treatment efficiency of 90%, is given on Figure 4-2. The orange areas on Figure 4-2 indicate the loadings that have been removed by the treatment process using the assumed treatment efficiency of 90%. The blue areas on Figure 4-2 indicate portions of the incoming loading which remain untreated. Assuming an estimated removal efficiency of 90%, a load reduction equivalent to 75% of the annual phosphorus loadings to Barr Lake can be achieved by treating inflows equal to or less than 133 cfs. This same process was used to evaluate the required diversion flow rates and annual volumes for each of the evaluated load reduction scenarios listed in Table 4-1 using the assumed phosphorus removal efficiencies listed in Table 4-2. In addition to the criteria discussed previously, several additional general overall criteria were applied to the evaluated treatment options. First, the system must be designed to minimize operational complexity and use gravity flow whenever possible. The evaluated options must also minimize area requirements and corresponding land purchase, whenever possible. Finally, the evaluated technology and system designs must incorporate reliable and proven treatment and operational technologies.

33 4-3 1, Phosphorus Loading (kg/day) Figure 4-1. Daily Phosphorus Loadings to Barr Lake from , Phosphorus Loading (kg/day) Treated Load (kg/day) Untreated (kg/day) Figure 4-2. Daily Phosphorus Loadings to Barr Lake to Achieve 75% Annual Total Phosphorus Reduction.

34 4-4 TABLE 4-2 ESTIMATED PROCESS REMOVAL EFFICIENCIES FOR THE ALUM TREATMENT AND MEDIA FILTRATION OPTIONS OPTION ASSUMED MEAN TOTAL PHOSPHORUS REMOVAL EFFICIENCY (%) Alum Treatment 90 Media Filtration 80 The simulation model was used to identify design criteria for treatment systems using alum and media filtration to achieve each of the four treatment options outlined on Table 4-1. A summary of treatment design criteria is given in Table 4-3. Information is provided on the maximum ditch discharge which must be diverted into a proposed treatment system for either alum or media filtration based upon the estimated process removal efficiencies summarized in Table 4-2. The corresponding average daily inflow to be treated is also provided. Calculations are also included for the average daily treated volume and average annual treated ditch inflow volume for each treatment process and treatment option. TABLE 4-3 SUMMARY OF TREATMENT DESIGN CRITERIA TREATMENT OPTION 1 (2,500 kg TP/yr) 2 (14,500 kg TP/yr) 3 (53,250 kg TP/yr) 4 (68,160 kg TP/yr) TREATMENT PROCESS MAXIMUM TREATED INFLOW RATE (cfs) AVERAGE DAILY TREATED INFLOW RATE (cfs) AVERAGE DAILY TREATED VOLUME (ac-ft) AVERAGE ANNUAL TREATED VOLUME (ac-ft) Alum ,931 Media Filtration ,185 Alum ,229 Media Filtration ,580 Alum ,549 Media Filtration ,261 Alum ,211 Media Filtration Cannot be achieved using media

35 4-5 For example, to achieve the 2,500 kg/yr of load reduction for total phosphorus outlined under Option 1, a process using alum treatment would divert all discharges through Burlington Ditch up to 3.7 cfs into the treatment system. This would correspond to an annual average diverted volume for treatment of 1,931 ac-ft. However, when using a media filtration system, inflows up to 4.2 cfs must be treated since the media filtration provides a lower mean annual phosphorus removal efficiency than alum. A media filtration treatment system for Option 1 would need to provide treatment for approximately 2,185 ac-ft/yr to be equivalent to an alum system which treated 1,931 ac-ft/yr. Similar calculations are provided for Option 2 (14,500 kg/yr total phosphorus) and Option 3 (53,250 kg/yr total phosphorus). However, the load reduction goal of 68,160 kg/yr of total phosphorus for Option 4, equivalent to a 96% annual load reduction, can only be approached using the alum treatment process. Although the removal efficiency for alum is conservatively estimated to be 90% for purposes of this evaluation, the actual removal efficiencies achieved during the laboratory jar testing ranged from 92-95% at the recommended optimum alum dose of 6 mg Al/liter. Therefore, it appears that alum treatment could approach, and possibly reach, the load reduction goals outlined under Option 4 to achieve the target TMDL reduction. However, this option would require that all discharges through the Burlington Ditch must receive alum treatment prior to discharge into Barr Lake. Under this option, the system must be capable of providing treatment for flows up to approximately 450 cfs which is the maximum inflow which occurred through the Burlington Ditch into Barr Lake over the period from However, the load reduction goals for Option 4 cannot be achieved using a media filtration process since the estimated removal efficiency for media filtration is 80%. 4.3 Proposed Treatment System Location The optimum location for a proposed treatment system for Barr Lake is near the point of inflow of the Burlington Ditch since all inputs into the ditch can be treated at a single location. If the proposed system is to operate by gravity, the ideal location would be immediately adjacent to the bifurcation structure so that the elevation difference upstream and downstream of the structure can be used as a driving force to move water through the treatment system. An overview of parcel ownership along the Burlington Ditch and in the vicinity of the bifurcation structure is given on Figure 4-3. FRICO owns much of the area immediately north of the bifurcation structure between the Burlington Ditch and Barr Lake. Another parcel west of the FRICO parcel is owned by the State of Colorado Division of Natural Resources. FRICO and the State of Colorado also own land around the southern and eastern perimeter of Barr Lake as well. The remaining parcels indicated on Figure 4-3 are currently in private ownership.

36 4-6 Figure 4-3. Parcel Ownership Along the Burlington Ditch and the Bifurcation Structure. Each of the proposed treatment options for Barr Lake will require land for construction of the treatment system components, although the quantity of land required varies significantly depending upon the selected option. Purchase of portions or all of the existing parcels along the Burlington Ditch may be required for implementation of the treatment system. A summary of current parcel listings in the vicinity of Barr Lake is given on Table 4-4 to provide estimates of existing land values in the general area. Six separate parcels are listed in Table 4-4, with three of the parcels currently zoned for agricultural purposes and the remaining three parcels zoned for agricultural/residential. Parcels zoned for agricultural land have unit prices ranging from $7,150-11,161/acre, while agricultural/residential properties range in value from 20,353-64,734/acre. The overall mean unit cost for the six parcels listed in Table 4-4 is slightly more than $25,000/acre. Therefore, for purposes of this evaluation, a land cost of $25,000/acre is included as a cost for land required for the evaluated treatment options.

37 4-7 LOCATION TABLE 4-4 CURRENT PARCEL LISTINGS IN THE VICINITY OF BARR LAKE (September 2014) PARCEL AREA (acres) PARCEL ZONING LISTED PRICE ($) UNIT PRICE ($/acre) Brighton 62.8 Agricultural 449,000 7,150 Brighton 2.91 Agricultural/Residential 125,000 42,955 Brighton 20 Agricultural 170,000 8,500 Henderson 2.07 Agricultural/Residential 134,000 64,734 Commerce City 36.8 Agricultural/Residential 749,000 20,353 Commerce City 1120 Agricultural 12,500,000 11,161 MEAN: $ 25, Alum Treatment of Ditch Inflows This section provides a discussion of conceptual system designs for alum inflow treatment systems to meet each of the evaluated load reduction scenarios for Barr Lake summarized in Table 4-1. Conceptual system designs are provided for each of the four load reduction options, including capital costs, annual O&M costs, and mass phosphorus removal costs Special Requirements The concept of using alum to treat inflows to lakes is a relatively new concept in the State of Colorado and would be subject to guidelines, restrictions, and water quality criteria established by the Colorado Water Quality Control Division (WQCD). A preliminary discussion was held by Ms. Laurie Rink and representatives of WQCD to review and discuss the proposed chemical treatment process and address anticipated permitting requirements. A summary of potential requirements and limitations for an alum inflow treatment system was prepared by Ms. Laurie Rink and summarized in correspondence dated August 19, 2014 to Mr. Dikar Kenan. Significant issues raised as part of this discussion are summarized as follows: 1. Treatment cannot occur in Waters of the State. Therefore, direct treatment of phosphorus in the Burlington Canal would be prohibited. 2. If flows were diverted out of the canal and treated in a side-stream, any water returned to the canal would be subject to a point source discharge permit. Of note: that permit would include effluent limitations for potentially dissolved aluminum (among other constituents) such that the in-stream (or in-canal) standard of 87 g/l would be met. Dilution from the canal could be considered in the calculation of the effluent limitation.

38 If some of the alum floc were allowed to pass back to the canal via the treated effluent and into Barr Lake, the 87 g/l would likely apply at the inlet to the lake since lakes are considered to have zero low flow and therefore no dilution. This assumption could be revised per results of a mixing study. The option of allowing floc to pass through to Barr Lake is based upon the notion that the floc may have additional capacity to bind phosphorus. The floc would be available to bind with phosphorus present in the lake s water column through sediment resuspension processes. Based upon the discussions outlined above, any alum treatment system proposed for the Burlington Ditch would have to be constructed so that the alum is added outside of the Burlington Ditch to prevent the alum addition from occurring in Waters of the State Basic Alum Treatment System Components A conceptual design of an off-line channel which provides for alum addition outside of Waters of the State is given on Figure 4-4. The illustrated channel is connected to the Burlington Ditch upstream from the existing water control structure by two 10-ft x 6-ft box culverts with sluice gates that can be opened or closed to regulate water flow through the off-line channel. The system would operate entirely by gravity, using the head difference between portions of the canal upstream and downstream of the existing control structure in Burlington Ditch. The proposed off-line channel would be approximately 50 ft in width to match the existing width of the Burlington Ditch and approximately 660 ft in length. The off-line channel would discharge downstream from the existing pedestrian bridge which crosses the ditch, and a new pedestrian bridge would be constructed to provide access across the off-line channel. The depth of the channel would be similar to the depth of the Burlington Ditch which is assumed to be approximately 8 ft. The total land area required for the off-line system is approximately 2.25 acres which includes a 50-ft buffer along the northern side. Flow measurements in the off-line channel would occur in a concrete-lined section approximately 30 ft in length and 50 ft in width. Flow monitoring would be conducted in this control section using a sensitive acoustic Doppler (AD) flow meter, with the flow signal returned to a control building (as indicated on Figure 4-4). Alum addition would occur at a location downstream from the point of flow measurement so that the turbulence created by the alum addition did not interfere with the flow measurements. The alum would be pre-mixed with water and injected into the flow through a series of nozzles installed on the bottom of the channel. A water intake structure and pumping system would be constructed upstream of the point of inflow for the off-line channel. The pumped canal water (generally about 300 gpm) will be used as carrier water to transport the alum to the point of alum addition in the channel. The additional volume created by the carrier water will provide mixing of the alum prior to being introduced into the channel under high pressure and velocity. The high velocity flow discharging from the injectors on the bottom of the channel will serve to mix the alum thoroughly with the water discharging through the off-line channel. Ditch water treated using this system would be discharged back into the Burlington Ditch, and the floc would be allowed to settle in Barr Lake.

39 4-9 Flow measurement Alum addition Off-line Channel Pedestrian bridge To Barr Lake Control Bldg. & chemical storage 2 10 x 6 Box Culverts with sluice gates 30 x 50 control section Rip-rap along channel banks Figure 4-4. Conceptual Design of an Off-Line Channel for Alum Addition. A conceptual cost estimate for the off-line channel illustrated on Figure 4-4 is given on Table 4-5. Conceptual costs are provided for land site work, excavation, rip-rap along the channel sides, two 10-ft x 6-ft box culverts with motorized sluice gates, the concrete channel control section, and the pedestrian bridge. The total estimated construction cost, including a 20% contingency, is $726,426. TABLE 4-5 CONCEPTUAL COST ESTIMATE FOR THE OFF-LINE CHANNEL OPTION NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 2.25 $ 25,000 $ 56, Site Work LS 1 15,000 15, Excavation CY 13, , Rip-Rap CY , ft x 6-ft Concrete Box Culvert LF 100 1, , ft x 6-ft Sluice Gate Motorized EA 2 32,000 64, Channel Control Section (30-ft x 50-ft x 6-inches) CY , Pedestrian Bridge EA 1 35,000 35,000 Sub-Total: $ 605,355 Contingency (20%) 121,071 TOTAL: $ 726,426

40 4-10 There are multiple additional components which are also common to each of the alum treatment options, regardless of the phosphorus load reduction goals. A conceptual schematic of the control building and chemical storage facility for the 2,500 kg TP/yr removal option is illustrated on Figure 4-5. The system consists of a pre-fabricated concrete control building which will contain the chemical metering pump, control panels, programmable logic controller (PLC), and flow meter readout and control system. This equipment is standard to each of the alum treatment options. The only potential modifications to this standard system is that multiple chemical metering pumps will be required for the larger phosphorus removal options to handle the range of alum additions due to the large variability in flow rates that must be treated. However, all remaining components and controls would be identical between the four systems. Each of the systems would also use a carrier water pumping system with an intake structure located in the Burlington Canal. The water pump station would be a wet well type facility with a submersible centrifugal pump that would provide a continuous, although adjustable, discharge of water from the canal to the point of alum addition. The alum would be pumped from the control building and injected into the carrier water flow so that the alum becomes mixed with the carrier water during movement through the piping system to the point of alum addition. The carrier water/alum mixture would be injected into the channel under high pressure to create turbulence that mixes the alum with the diverted ditch water. Each of the alum treatment options would also require an alum storage tank with a 12,500-gallon tank proposed for the 2,500 kg TP/yr option. A conceptual schematic of alum treatment system components for the 14,500 kg TP/yr removal option is given on Figure 4-6. The basic system components are identical to the 2,500 kg TP/yr option with the exception that two chemical storage tanks are included rather than one. A conceptual schematic of alum treatment system components for the 53,250 kg TP/yr and 68,160 kg TP/yr removal options is given on Figure 4-7. The basic system components are identical to the components required for the 2,500 kg TP/yr and 14,500 kg TP/yr removal options with the exception that four chemical storage tanks are provided for the larger removal options. In addition, both low-end and high-end chemical metering pumps would be required for this option to provide accurate alum addition over the wide range of anticipated flow rates to be treated Design Parameter Summary A summary of basic alum treatment system design parameters for the evaluated load reduction options is given on Table 4-6. Information is provided for maximum treated canal discharges, average daily treated inflow rates and volumes, annual treated volume, the recommended alum addition dose, annual alum use, mean and maximum daily alum use, annual alum cost, and floc production. The annual treated inflow volumes range from 1,931 ac-ft/yr for the 2,500 kg TP/yr removal option to 45,211 ac-ft/yr for the 96% removal (68,160 kg TP/yr) option. Annual alum use ranges from 64,440 gallons to 1,508,757 gallons for the evaluated options. At an assumed alum cost of $0.95/gallon, the estimated annual alum cost for phosphorus removal in Burlington Ditch ranges from $61,218-1,433,319. The treatment options will generate ac-ft/yr of wet floc, corresponding to ac-ft/yr of dry floc, based upon an assumed moisture content of 95%.

41 4-11 Piping to point of alum addition 1 12,500 gal. tank Insulated metal building Control building (pre-fabricated concrete) Valves Alum addition to carrier water Flow meter conduit Water pump station HDPE intake piping to pump station 2 10 x 6 Box Culverts with sluice gates Carrier water intake structure Figure 4-5. Conceptual Schematic of Alum Treatment System Components for the 2,500 kg TP/yr Removal Option.

42 4-12 Piping to point of alum addition 2 12,500 gal. tanks Insulated metal building Control building (pre-fabricated concrete) Valves Alum addition to carrier water Flow meter conduit Water pump station HDPE intake piping to pump station 2 10 x 6 Box Culverts with sluice gates Carrier water intake structure Figure 4-6. Conceptual Schematic of Alum Treatment System Components for the 14,500 kg TP/yr Removal Option.

43 ,500 gal. tanks Insulated metal building Piping to point of alum addition Control building (pre-fabricated concrete) Valves Alum addition to carrier water Flow meter conduit Water pump station HDPE intake piping to pump station 2 10 x 6 Box Culverts with sluice gates Carrier water intake structure Figure 4-7. Conceptual Schematic of Alum Treatment System Components for the 53,250 and 68,160 kg TP/yr Removal Options.

44 4-14 TABLE 4-6 ALUM INFLOW TREATMENT DESIGN PARAMETERS FOR THE EVALUATED LOAD REDUCTION OPTIONS OPTION PARAMETER UNITS 2,500 kg TP/yr 14,500 kg TP/yr 53,250 kg TP/yr 68,160 kg TP/yr Maximum Treated Inflow cfs Average Daily Treated Inflow cfs ac-ft/day Annual Treated Volume ac-ft/yr 1,931 11,229 37,549 45,211 Alum Addition Dose mg Al/liter Annual Alum Use gallons 64, ,728 1,253,065 1,508,757 Mean Daily Alum Use (62.4 cfs) gallons 179 1,029 3,432 4,130 Maximum Daily Alum Use (450 cfs) gallons 245 1,622 8,803 29,786 Alum Cost $/gallon $ , , ,190, ,433,319 Wet Floc Production 1 % of flow Wet Floc Generation ac-ft Dry Floc Generation ac-ft Assumes an additional 25% consolidation Conceptual Cost Estimates for Basic System Components Conceptual cost estimates for basic alum treatment components for the evaluated load reduction options are provided on Table 4-7. Each of the cost estimates includes site work; the pre-fabricated concrete control building; pumps, flow meters, and instrumentation; chemical storage tanks; tank pad and containment area with climate controlled cover; electrical and mechanical service; a carrier water pump station and intake structure; piping; inlet; and mobilization. Each of the cost estimates also includes a 20% contingency. The estimated conceptual construction cost for the 2,500 kg TP/yr removal option is approximately $435,030. The removal option of 14,500 kg TP/yr adds an additional storage tank and an increased area for the tank pad and containment as well as the climate controlled tank cover. The estimated construction cost for this option is $484,830. The construction cost estimate for the 53,250 kg TP/yr and 68,160 kg TP/yr removal options are identical since the same system components will be required for either option. This option includes four separate alum storage tanks, along with a larger climate controlled tank pad and containment area. An additional cost has also been added for an additional alum metering pump to cover the wide range of anticipated discharges to be treated. The estimated total construction cost for this option is $608,430 which includes a 20% contingency.

45 4-15 TABLE 4-7 CONCEPTUAL COST ESTIMATES FOR BASIC ALUM TREATMENT COMPONENTS FOR THE EVALUATED LOAD REDUCTION OPTIONS 2,500 kg TP/yr Option NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 0.5 $ 25,000 $ 12, Site Work LS 1 15,000 15, Pre-fabricated Concrete Pump/Control Building with Gravel Base EA 1 18,000 18, Pumps, Flow Meter, and Instrumentation LS 1 125, , Alum Storage Tanks (12,500 gallons each) with Piping EA 1 17,500 17, Tank Pad and Containment Area (20-ft x 20-ft) YDS , Pre-fabricated Climate Controlled Aluminum Cover for Tank Pad LS 1 15,000 15, Electrical Service, Conduits, Wiring, etc. LS 1 30,000 30, Alum Addition Section (25-ft x 5-ft x 12-inches) LS 1 5,000 5, Carrier Water Pump Station and Piping LS 1 40,000 40, Carrier Water Intake Structure in Burlington Ditch LS 1 10,000 10, Piping from Intake to Carrier Pump Station (24-inch HDPE) FT , Type D Inlet EA 1 5,000 5, Mobilization LS 1 50,000 50,000 Sub-Total: $ 362,525 Contingency (20%) 72,505 TOTAL: $ 435,030 14,500 kg TP/yr Option NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 0.5 $ 25,000 $ 12, Site Work LS 1 15,000 15, Pre-fabricated Concrete Pump/Control Building with Gravel Base EA 1 18,000 18, Pumps, Flow Meter, and Instrumentation LS 1 125, , Alum Storage Tanks (12,500 gallons each) with Piping EA 2 17,500 35, Tank Pad and Containment Area (20-ft x 40-ft) YDS , Pre-fabricated Climate Controlled Aluminum Cover for Tank Pad LS 1 25,000 25, Electrical Service, Conduits, Wiring, etc. LS 1 30,000 30, Alum Addition Section (25-ft x 5-ft x 12-inches) LS 1 5,000 5, Carrier Water Pump Station and Piping LS 1 40,000 40, Carrier Water Intake Structure in Burlington Ditch LS 1 10,000 10, Piping from Intake to Carrier Pump Station (24-inch HDPE) FT , Type D Inlet EA 1 5,000 5, Mobilization LS 1 50,000 50,000 Sub-Total: $ 404,025 Contingency (20%) 80,805 TOTAL: $ 484,830

46 4-16 TABLE CONTINUED CONCEPTUAL COST ESTIMATES FOR BASIC ALUM TREATMENT COMPONENTS FOR THE EVALUATED LOAD REDUCTION OPTIONS 53,230 kg TP/yr and 68,160 kg TP/yr Options NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 0.5 $ 25,000 $ 12, Site Work LS 1 15,000 15, Pre-fabricated Concrete Pump/Control Building with Gravel Base EA 1 18,000 18, Pumps, Flow Meter, and Instrumentation LS 1 140, , Alum Storage Tanks (12,500 gallons each) with Piping EA 4 17,500 70, Tank Pad and Containment Area (40-ft x 40-ft) YDS , Pre-fabricated Climate Controlled Aluminum Cover for Tank Pad LS 1 50,000 50, Electrical Service, Conduits, Wiring, etc. LS 1 30,000 30, Alum Addition Section (25-ft x 5-ft x 12-inches) LS 1 5,000 5, Carrier Water Pump Station and Piping LS 1 40,000 40, Carrier Water Intake Structure in Burlington Ditch LS 1 10,000 10, Piping from Intake to Carrier Pump Station (24-inch HDPE) FT , Type D Inlet EA 1 5,000 5, Mobilization LS 1 50,000 50,000 Sub-Total: $ 507,025 Contingency (20%) 101,405 TOTAL: $ 608, Floc Removal Options Three separate sub-options were evaluated for each of the alum treatment load reduction alternatives based upon different methods of collection and disposal of the alum floc. The first option involves treating the ditch water in an off-line channel, similar to the channel illustrated on Figure 4-4, with the generated floc discharging into Barr Lake and accumulating within the sediments. The accumulated alum floc will provide a substantial secondary benefit of inactivating sediment phosphorus release in areas of floc accumulation. The second floc management option is referred to as the Partial Floc Removal Option which involves collecting a portion of the generated floc in a dedicated settling basin, with the remaining floc allowed to discharge to Barr Lake. This option is designed to reduce the volume of accumulated floc in Barr Lake while still allowing sufficient floc accumulation to provide significant sediment inactivation. For purposes of this evaluation, it is assumed that this option provides for collection of 50% of the annual floc volume, with the remainder discharging to Barr Lake.

47 4-17 The final floc management option, referred to as the Full Floc Collection Option, involves full collection of all generated floc in a dedicated floc collection basin sufficient in size to store a minimum of one year s worth of floc accumulation for the particular design treatment option. However, even with a treatment objective designed to capture all of the collected floc, it still appears reasonable that conditions could occur under which at least portions of the annual alum treated flow could potentially bypass the floc settling area, allowing floc to discharge to Barr Lake. Therefore, the off-line channel illustrated on Figure 4-4 is assumed to be a standard component for each of the floc collection options, including the Full Floc Collection Option. A series of evaluations was conducted to identify preliminary design parameters for floc collection ponds for each of the four load reduction options. This analysis was conducted for both the partial floc removal option and full floc removal option. Each of the preliminary designs is based upon the wet floc generation volumes summarized near the bottom of Table 4-6 for each load reduction option. A schematic of a typical settling pond design for treatment of inflows from the Burlington Ditch is given on Figure 4-8. The design of the settling pond is determined in large part by the physical and hydraulic characteristics of the ditch. In each of the proposed floc collection system designs, water is diverted from the Burlington Ditch, alum is added, and the treated water and floc are discharged into a settling basin. To allow inflows into the off-line floc collection system even during low water conditions, the bottom of the off-line channel leading to the floc settling pond needs to be similar to the bottom elevation of the Burlington Ditch. The floc storage volume is then created at elevations below the ditch bottom so that the floc accumulation will not hinder water movement through the pond. The volume of the pond below the ditch bottom elevation is designed to accommodate a full year of floc storage based upon the wet floc generation rates summarized in Table 4-6. This evaluation assumes that the depth of the floc storage portion of the pond will be approximately 4 ft which is a compromise between expanding floc storage at deeper excavation depths and providing a floc depth which is shallow enough that it could be dried and desiccated during a single dry season period. Therefore, each of the floc storage pond options assumes a 4-ft depth for floc storage. Pond Water Level (Burlington Ditch water level) ~ 5 ft. Settling Zone (Minimum detention time = 1.5 hour Volume = Max inflow for 1.5 hour) ~ 6 ft. Floc Storage Floc Storage (1 year of floc accumulation) (Volume = annual floc generation) 4 ft. Bottom elev. of Burlington Ditch Figure 4-8. Typical Settling Pond Design for the Barr Lake Inflow.

48 4-18 The water level above the floc storage volume will be regulated by water levels in the Burlington Ditch. This analysis assumes that the pond water level will be approximately 6-7 ft in depth above the floc storage area, based upon a hydrologic equilibrium between water levels upstream and downstream of the bifurcation structure. The analysis also assumes that the Burlington Ditch water level is approximately 5 ft below the existing ground level elevation in the area where the floc pond is proposed. Therefore, the total depth of the excavation required will include the 4 ft of floc storage plus 6 ft free water plus an additional 5 ft of overburden above the anticipated pond level, for a total assumed excavation depth of 15 ft. A summary of pond design parameters for the full floc collection option is given in Table 4-8 for each of the four load reduction scenarios. The required area of the floc pond is calculated for each load reduction option by dividing the annual wet floc generation rates (summarized on Table 4-6) by the assumed maximum floc depth of 4 ft. This process results in estimated pond surface areas ranging from 1.1 acres for the 2,500 kg TP/yr option to 25.5 acres for the 68,160 kg TP/yr option. TABLE 4-8 POND DESIGN PARAMETERS FOR THE FULL FLOC COLLECTION OPTIONS PARAMETER UNITS 2,500 kg TP/yr 14,500 kg TP/yr OPTION 53,250 kg TP/yr 68,160 kg TP/yr Required Floc Pond Area acres Annual Floc Accumulation in/yr ft/yr Minimum Floc Settling Time hours Minimum Settling Zone Requirements ac-ft ft Minimum Required Pond Volume 1 ac-ft yd 3 7,750 45, , ,199 Assumed Pond Depth 1 ft Assumed Overburden Depth to Water Level ft Total Excavation Depth ft Excavated Pond Volume ac-ft yd 3 26, , , , Below pond water level

49 4-19 In addition to floc storage, the floc settling pond must also provide a minimum detention time of 1.5 hours for the incoming alum treated water to ensure adequate opportunity for floc settling before the water discharges from the pond. Estimates of the minimum required water volume were calculated for each treatment option by multiplying the maximum design treated inflow rates (summarized on Table 4-6) times 1.5 hours to obtain the required settling zone volume. This volume is then added to the floc storage volume to obtain the minimum required pond volume below ambient water level elevations. The designed pond volume must equal or exceed this value for acceptable pond performance. However, as indicated on Table 4-8, the water depth requirements for the aqueous settling zone portion of the pond ranges from 0.4 ft for the 2,500 kg TP/yr treatment option to 2.4 ft for the 68,160 kg TP/yr option. These values are substantially lower than the anticipated water level within the pond of 6-7 ft based upon the Burlington Ditch water level elevation. Therefore, each of the proposed settling pond designs will easily meet both the required floc storage and settling zone volumes. As mentioned previously, the pond depth is assumed to be approximately 10 ft below the pond water level, with the bottom 4 ft allocated to floc storage and a 6-ft water depth assumed above the floc storage within the pond. The existing overburden in the area proposed for the floc collection ponds is approximately 5 ft to reach the anticipated Burlington Ditch water level. Therefore, the total excavation depth assumed for a floc settling pond is 15 ft. The anticipated excavation volume for the pond is obtained by multiplying the required floc pond volume times the total excavation depth of 15 ft. This process results in anticipated excavation volumes ranging from 16.4 ac-ft for the 2,500 kg TP/yr option to 383 ac-ft for the 68,160 kg TP/yr option. Equivalent volumes in terms of yd 3 are also provided for each of these options. The estimated volumes are used to generate estimates of the potential construction costs for floc settling ponds designed for each load reduction option. An evaluation of anticipated pond design characteristics for the partial floc collection option was also conducted, and a summary of the estimated design parameters is given in Table 4-9. This analysis assumes a floc capture rate of approximately 50%, with 50% of the floc collected in a dedicated floc settling pond and the remaining 50% allowed to discharge into Barr Lake. Since only half of the annual generated floc is proposed to be collected, the required floc pond area requirements for the partial floc collection option are one-half of the required floc pond requirements for the full floc collection option. These ponds would also be constructed to a depth of 15 ft to meet the assumptions regarding floc storage, pond water levels, and overburden depth, so the anticipated excavated pond volume for the partial floc collection option would be half of the anticipated pond volume for the full floc collection option. An evaluation was also conducted to examine impacts of discharge of alum floc directly into Barr Lake. This analysis was conducted for two sub-options which include full discharge of all generated floc to Barr Lake as well as introduction of 50% of the generated floc to Barr Lake. Depositional areas for alum floc in Barr Lake depend heavily upon water level elevations at the time of the floc inputs. An illustration of floc deposition into Barr Lake during low water conditions is given on Figure 4-9. Under these conditions, the channelized portion of the Burlington Canal extends far into Barr Lake, and the alum floc will be deposited into deeper central portions of the lake. However, under high water conditions, illustrated on Figure 4-10, the floc will be introduced primarily into southwestern portions of Barr Lake although smaller portions of the floc may migrate farther into the lake.

50 4-20 TABLE 4-9 POND DESIGN PARAMETERS FOR THE PARTIAL FLOC COLLECTION OPTIONS PARAMETER UNITS 2,500 kg TP/yr 14,500 kg TP/yr OPTION 53,250 kg TP/yr 68,160 kg TP/yr Fraction of Floc Captured % Required Floc Pond Area acres Floc deposited into deeper areas of the lake Figure 4-9. Floc Discharging into Barr Lake During Low Water Conditions.

51 4-21 Floc deposited into shoreline and deep areas of the lake Figure Floc Discharging into Barr Lake During High Water Conditions. Based upon the typical high and low water conditions in Barr Lake, areas of assumed annual floc deposition in Barr Lake are illustrated on Figure Approximately 75% of the floc introduced into the lake is expected to accumulate in central and southwestern portions of the lake, with approximately 25% of the floc inputs being carried farther into Barr Lake. For purposes of this analysis, the primary settling area is assumed to occupy an area of 300 acres. A summary of estimated floc accumulation rates in Barr Lake is given on Table This analysis is conducted for the primary floc settling area only since this area accumulates the largest portion of the annual floc inputs (75%). This analysis is conducted for the option where all floc is introduced into Barr Lake, as well as the partial floc collection option based on an assumption that 50% of the floc loading is removed prior to discharge to Barr Lake.

52 4-22 Secondary Area ~ 25% Primary Area ~ 75% Figure Assumed Annual Floc Depositional Areas in Barr Lake. TABLE 4-10 SUMMARY OF IN-LAKE FLOC ACCUMULATION IN BARR LAKE FOR THE FULL AND PARTIAL FLOC DISCHARGE OPTIONS PARAMETER UNITS 2,500 kg TP/yr 14,500 kg TP/yr OPTION 53,250 kg TP/yr 68,160 kg TP/yr Assumed Floc Settling Area in Lake acres Full Floc Discharge Partial Floc Discharge Annual Floc Volume Discharged to Lake ac-ft Annual Floc Accumulation in Lake Floc Volume Discharged to Lake Annual Floc Accumulation in Primary Settling Area in Lake in/yr wet in/yr dry % of total ac-fat in/yr wet in/yr dry

53 4-23 As indicated on Table 4-10, introduction of all of the generated alum floc into Barr Lake would result in wet alum floc accumulation rates, within the assumed 300-acre primary floc settling area, ranging from 0.13 in/yr for the 2,500 kg TP/yr option to 3.1 in/yr for the 68,160 kg TP/yr option. If this floc were allowed to desiccate under periods of low water conditions, the resulting accumulation rates would range from less than 0.01 in/yr to 0.2 in/yr for the range of evaluated phosphorus load reduction options. Annual floc accumulations in Barr Lake for the partial floc discharge are provided near the bottom of Table 4-10 based upon an assumed floc discharge equivalent to 50% of the annual generated volume, with 75% of the generated floc settling in the primary 300-acre settling zone. Floc accumulation within the assumed 300-acre primary settling area would range from in/yr for the evaluated phosphorus removal options. If this floc volume were allowed to desiccate and dry within Barr Lake under low water conditions, the annual floc accumulations would be reduced to depths ranging from in/yr within the 300-acre primary floc settling area Conceptual Floc Collection Designs Evaluations were conducted for the four separate load reduction options, each of which has a full and partial floc collection sub-option, for a total of eight potential conceptual designs for floc collection. Detailed evaluations for floc removal systems were developed for the 2,500 kg TP/yr and 14,500 kg TP/yr removal options. As indicated on Table 4-8, required floc pond areas for collection of the full floc generated from alum treatment systems to achieve the desired phosphorus reduction goals are 1.1 acres for the 2,500 kg TP/yr option and 6.3 acres for the 14,500 kg TP/yr option. Each of these conceptual pond designs can fit easily into the existing land currently owned by FRICO adjacent to the bifurcation structure. Floc collection areas for the 53,250 kg TP/yr and 68,160 kg TP/yr removal options have minimum surface area requirements of 21 acres and 25.5 acres, respectively. These areas substantially exceed the land currently in possession of FRICO and would require purchase of multiple adjacent parcels for construction. The specific configuration of floc removal systems for these load reduction options would depend upon the availability of land adjacent to the site. As a result, conceptual schematic drawings for floc removal systems are not provided for the 53,250 kg TP/yr or 68,160 kg TP/yr removal options, although detailed cost estimates are provided based upon anticipated excavation volumes and system requirements. A schematic floc removal system for full floc removal with the 2,500 kg TP/yr treatment option is given on Figure The system illustrated on Figure 4-12 runs parallel to the off-line channel for direct alum treatment. Water would be introduced into the system through a 36-inch HDPE pipe with an invert elevation near the bottom of the Burlington Canal. The inflow water would be carried through a 30-ft wide inflow channel, with alum addition occurring approximately mid-way down the channel using the same method outlined previously for the offline channel. The alum treated water would migrate into the 1.1-acre floc settling pond, with the clarified water discharging through a 36-inch HDPE into a 30-ft wide outfall channel. A reconfigured and expanded pedestrian bridge is also illustrated on Figure 4-12 which would be necessary to cross the combined channels from the floc settling pond and the off-line channel.

54 Inflow Canal 36 HDPE with sluice gate 1.1 ac. Floc Settling Pond 36 HDPE with sluice gate Rip-rap 30 Outflow Canal Rip-rap Alum Addition Pedestrian Bridge Rip-rap Flow Measurement 36 HDPE Figure Schematic of Floc Removal System for Full Floc Removal with the 2,500 kg TP/yr Treatment Option. A schematic of a proposed floc removal system for the 14,500 kg TP/yr removal option is given on Figure This option requires a 6.3-acre floc settling pond which will easily fit within the existing FRICO property. Water would be introduced into an inflow channel by a 4-ft x 6-ft concrete box culvert (CBC). Flow measurement would occur at the end of the box culvert, with the alum addition occurring downstream of flow measurement. The alum treated water would be introduced into the pond through two additional 4-ft x 6-ft CBCs with sluice gates to regulate discharge rates through the pond. The alum treated water would migrate through the pond and discharge through two additional 4-ft x 6-ft CBCs with sluice gates, entering the outfall channel and rejoining the Burlington Canal to discharge into Barr Lake. An expanded pedestrian bridge is also illustrated for this option.

55 Inflow Canal 4 x 6 CBC with sluice gate 6.3 ac. Floc Settling Pond 4 x 6 CBC with sluice gate Rip-rap 30 Outflow Canal Rip-rap Alum Addition Pedestrian Bridge Rip-rap Flow Measurement 4 x 6 CBC Figure Schematic of Floc Removal System for Full Floc Removal with the 14,500 kg TP/yr Treatment Option. A tabular summary of conceptual construction costs for the floc collection systems associated with the 2,500 kg TP/yr removal option is given in Table Separate cost estimates are provided for full floc collection in an off-line pond along with partial (50%) floc collection. The cost estimates summarized in Table 4-11 are based upon the design floc collection pond parameters summarized in Table 4-8 for the full collection option and in Table 4-9 for the partial floc collection option. Land costs for each of the evaluated options are assumed to be $25,000/acre. The land requirement is based upon the required floc pond areas for full and partial floc collection (summarized in Table 4-8 and 4-9), with an additional 25% allowance for roadways and maintenance berms. Excavation costs for the full floc collection system are based upon actual quantities obtained from the schematic floc removal system summarized on Figure Additional fixed costs are provided for rip-rap, piping, motorized sluice gates, a channel control section for flow measurement, and a pedestrian bridge. The total estimated conceptual construction cost for a full off-line floc collection system with the 2,500 kg TP/yr option is $615,018, which includes a 20% contingency.

56 4-26 TABLE 4-11 CONCEPTUAL CONSTRUCTION COSTS FOR FLOC COLLECTION WITH THE 2,500 kg TP/yr REMOVAL OPTION Option A: Full Floc Collection in Off-Line Pond NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 0.5 $ 25,000 $ 12, Site Work LS 1 15,000 15, Excavation a. Settling Pond b. Inflow Canal c. Outflow Canal CY CY CY 16, ,389 18,387 19, Rip-Rap CY , inch HDPE LF , inch Sluice Gate (Motorized) EA 2 8,000 16, Channel Control Section (30-ft x 30-ft x 6-in) CY , Pedestrian Bridge EA 1 35,000 35,000 Sub-Total: $ 512,515 Contingency (20%) 102,503 TOTAL: $ 615,018 Option B: Partial (50%) Floc Collection in Off-Line Pond NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 0.7 $ 25,000 $ 17, Site Work LS 1 10,000 10, Excavation a. Settling Pond b. Inflow Canal c. Outflow Canal CY CY CY 8, ,194 12,258 13, Rip-Rap CY , inch HDPE LF , inch Sluice Gate (Motorized) EA 2 8,000 16, Channel Control Section (30-ft x 30-ft x 6-in) CY , Pedestrian Bridge EA 1 35,000 35,000 Sub-Total: $ 313,791 Contingency (20%) 62,758 TOTAL: $ 376,549

57 4-27 A cost estimate is also provided in Table 4-11 for the partial floc collection option. The required land area is based upon the required floc area (summarized on Table 4-9) with an additional 25% allowance for maintenance areas and roads. The excavation quantity for the settling pond is assumed to be half of the excavation quantity for the full floc collection pond, with the inflow canal and outflow canal volumes assumed to be two-thirds of the full floc collection pond design. Costs for rip-rap, piping, motorized sluice gate, and channel control section are assumed to be the same as the full floc collection system. Based upon this analysis, the estimated conceptual construction cost for the off-line partial floc collection system is $376,549. Excavation costs are assumed at $20/yd 3 for both the full and partial floc collection cost estimates. A conceptual cost estimate for floc removal options associated with the 14,500 kg TP/yr removal option is given in Table The quantity summarized in this table for the full floc collection option are based upon actual quantities associated with the conceptual system summarized in Figure Pond excavation costs are assumed to be $20/yd 3 for both the full and partial floc collection cost estimates. Quantity estimates for the partial floc collection system were generated in the manner previously described for the 2,500 kg TP/yr system. The estimated capital construction costs for the 14,500 kg TP/yr removal option is $4,001,402 for the full floc collection system, and $2,158,724 for the partial floc collection system. A tabular summary of conceptual construction costs for full and partial floc collection systems associated with the 53,250 kg TP/yr removal option is given on Table Land requirements for the full and partial floc collection options are based upon the required floc pond areas summarized in Tables 4-8 and 4-9, with an additional 25% allowance for maintenance areas and roadways. The excavation volume is calculated by multiplying the required pond size times a depth of 15 ft. Excavation costs for this option have been reduced from $20/yd 3 to $15/yd 3 due to the larger volume of material which will likely generate a lower bid cost. Cost estimates are also included for box culverts, sluice gates, channel control section, and pedestrian bridge. The quantity excavation estimates for the partial floc collection system are based upon the same methods described for the previous options. Overall, the estimated construction cost for the off-line pond for full floc collection of the 53,250 kg TP/yr removal option is $10,750,505, with an estimated construction cost of $5,688,131 for the partial floc collection system. A tabular summary of conceptual cost estimates for full and partial floc collection associated with the 68,160 kg TP/yr removal option is given on Table Estimates provided in this table were calculated similar to methods used for the previous options. This option also uses a reduced excavation rate of $15/yd 3. The total estimated construction cost of the 25.5-acre floc settling pond for full floc collection with this option is $12,870,983, with an estimated construction cost of $6,762,833 for the partial floc collection system.

58 4-28 TABLE 4-12 CONCEPTUAL CONSTRUCTION COSTS FOR FLOC COLLECTION WITH THE 14,500 kg TP/yr REMOVAL OPTION Option A: Full Floc Collection in Off-Line Pond NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 7.9 $ 25,000 $ 196, Site Work LS 1 65,000 65, Excavation a. Settling Pond b. Inflow Canal c. Outflow Canal CY CY CY 135,461 3,188 2, ,709,220 63,760 47, Rip-Rap CY , ft x 6-ft Concrete Box Culvert LF , ft x 6-ft Sluice Gate (motorized) EA 4 16,000 64, Channel Control Section (30-ft x 30-ft x 6-in) CY , Pedestrian Bridge EA 1 35,000 35,000 Sub-Total: $ 3,334,502 Contingency (20%) 666,900 TOTAL: $ 4,001,402 Option B: Partial (50%) Floc Collection in Off-Line Pond NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 4.0 $ 25,000 $ 100, Site Work LS 1 30,000 30, Excavation a. Settling Pond b. Inflow Canal c. Outflow Canal CY CY CY 67,731 2,125 1, ,354,610 42,507 31, Rip-Rap CY , ft x 6-ft Concrete Box Culvert LF , ft x 6-ft Sluice Gate (motorized) EA 2 16,000 32, Channel Control Section (30-ft x 30-ft x 6-in) CY , Pedestrian Bridge EA 1 35,000 35,000 Sub-Total: $ 1,798,937 Contingency (20%) 359,787 TOTAL: $ 2,158,724

59 4-29 TABLE 4-13 CONCEPTUAL CONSTRUCTION COSTS FOR FLOC COLLECTION WITH THE 53,250 kg TP/yr REMOVAL OPTION Option A: Full Floc Collection in Off-Line Pond NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 26.3 $ 25,000 $ 656, Site Work LS 1 90,000 90, Excavation a. Settling Pond b. Inflow Canal c. Outflow Canal CY CY CY 508,200 5,241 2, ,623,000 78,615 38, Rip-Rap CY 1, , ft x 6-ft Concrete Box Culvert LF 200 1, , ft x 6-ft Sluice Gate (motorized) EA 4 32, , Channel Control Section (50-ft x 30-ft x 6-in) CY , Pedestrian Bridge EA 1 35,000 35,000 Sub-Total: $ 8, Contingency (20%) 1,791,751 TOTAL: $ 10,750,505 Option B: Partial (50%) Floc Collection in Off-Line Pond NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 13.1 $ 25,000 $ 656, Site Work LS 1 70,000 70, Excavation a. Settling Pond b. Inflow Canal c. Outflow Canal CY CY CY 254,100 3,494 1, ,811,500 52,410 25, Rip-Rap CY 1, , ft x 6-ft Concrete Box Culvert LF 180 1, , ft x 6-ft Sluice Gate (motorized) EA 4 32, , Channel Control Section (50-ft x 30-ft x 6-in) CY , Pedestrian Bridge EA 1 35,000 35,000 Sub-Total: $ 4,740,109 Contingency (20%) 948,022 TOTAL: $ 5,688,131

60 4-30 TABLE 4-14 CONCEPTUAL CONSTRUCTION COSTS FOR FLOC COLLECTION WITH THE 68,160 kg TP/yr REMOVAL OPTION Option A: Full Floc Collection in Off-Line Pond NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 31.9 $ 25,000 $ 796, Site Work LS 1 50,000 50, Excavation a. Settling Pond b. Inflow Canal c. Outflow Canal CY CY CY 617,100 6,000 3, ,256,500 90,000 45, Rip-Rap CY 1, , ft x 6-ft Concrete Box Culvert LF 200 1, , ft x 6-ft Sluice Gate (motorized) EA 4 32, , Channel Control Section (50-ft x 30-ft x 6-in) CY , Pedestrian Bridge EA 1 35,000 35,000 Sub-Total: $ 10,725,819 Contingency (20%) 2,145,164 TOTAL: $ 12,870,983 Option B: Partial (50%) Floc Collection in Off-Line Pond NO. ITEM DESCRIPTION UNITS QUANTITY UNIT COST PRICE ($) 1. Land ACRES 16.0 $ 25,000 $ 400, Site Work LS 1 50,000 50, Excavation a. Settling Pond b. Inflow Canal c. Outflow Canal CY CY CY 308,550 4,000 2, ,628,250 60,000 30, Rip-Rap CY 1, , ft x 6-ft Concrete Box Culvert LF 180 1, , ft x 6-ft Sluice Gate (motorized) EA 4 32, , Channel Control Section (50-ft x 30-ft x 6-in) CY , Pedestrian Bridge EA 1 35,000 35,000 Sub-Total: $ 5,635,694 Contingency (20%) 1,127,139 TOTAL: $ 6,762,833

61 Annual O&M Costs A tabular summary of estimated annual O&M costs for each of the evaluated alum inflow treatment options is given on Table Estimates of annual alum usage for each of the treatment options is based upon information provided in Table 4-6. The unit cost for alum in the Denver area is assumed to be $0.95/gallon based upon a cost estimate projection provided by ChemTrade Logistics for January 1, Labor requirements are variable depending upon the complexity of the treatment process and range from 15 hours/week for the direct floc discharge options to 32 hours/week for the larger treatment options. Labor costs are assumed to be $30/hour. O&M costs are included for utilities at an assumed rate of approximately $2,000/month. Repairs and allowance for repairs and miscellaneous items are also provided for each treatment option. For systems involving floc removal, the estimated floc removal costs are based upon the dry floc generation volumes for the various treatment options (summarized in Table 4-6). Annual O&M costs for the evaluated alum treatment options range from $133,618/yr for direct floc discharge with the 2,500 kg TP/yr removal option to $1,746,355/yr for full floc capture with the 68,160 kg TP/yr removal option. The information summarized in Table 4-15 is used in a subsequent section to estimate phosphorus removal costs Phosphorus Removal Costs Phosphorus removal costs were calculated for each of the four load reduction treatment options and the three floc disposal options. The phosphorus removal costs are based upon a 20- year life cycle analysis with an assumed interest rate of 4%. The 20-year life cycle cost is calculated by adding the estimated construction cost plus 20 years of annual O&M cost which is converted to a present worth cost. The resulting 20-year present worth cost is then divided by 20 years worth of phosphorus load reductions to obtain the estimated phosphorus mass removal cost in dollars/kg. A summary of calculated phosphorus removal costs for the 2,500 kg TP/yr removal option is given in Table For direct floc discharge to the lake, the construction cost is calculated by adding the basic system cost for this option (summarized in Table 4-7) plus the estimated cost for the off-line channel (summarized in Table 4-5). Estimates of annual O&M costs are obtained from information provided in Table The estimated phosphorus removal cost for direct discharge of floc to the lake is $59.50/kg. Construction costs for the full floc capture option with the 2,500 kg TP/yr removal scenario are equal to the construction cost assumed for the direct floc discharge plus the cost of the floc settling area for full floc capture. Estimates of annual O&M costs for this option are obtained from the information provided in Table Present worth phosphorus removal cost for the full floc capture option is approximately $75.90/kg. The phosphorus removal cost for the partial floc capture option is obtained by adding the construction cost for direct floc discharge to the cost of the floc collection system for partial floc capture (summarized in Table 4-11). The estimated annual O&M costs are obtained from information provided in Table The overall 20-year present worth phosphorus removal cost for this option is $70.20/kg.

62 4-32 TABLE 4-15 ESTIMATED ANNUAL O&M COSTS FOR THE ALUM INFLOW TREATMENT OPTIONS Floc Discharge to Lake OPTION 2,500 kg TP/yr Removal Option UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 64, ,218 b. Labor (15 hours/week) man-hours ,400 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 Annual Total: $ 133,618 Full Floc Capture OPTION UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 64, ,218 b. Labor (20 hours/week) man-hours 1, ,200 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 e. Floc Removal (dry) CY ,010 Annual Total: $ 148,428 Partial Floc Capture OPTION UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 64, ,218 b. Labor (20 hours/week) man-hours 1, ,200 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 e. Floc Removal (dry) CY ,505 Annual Total: $ 144,923

63 4-33 TABLE CONTINUED ESTIMATED ANNUAL O&M COSTS FOR THE ALUM INFLOW TREATMENT OPTIONS Floc Discharge to Lake OPTION 14,500 kg TP/yr Removal Option UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 374, ,992 b. Labor (15 hours/week) man-hours ,400 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 Annual Total: $ 443,392 Full Floc Capture OPTION UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 374, ,992 b. Labor (20 hours/week) man-hours 1, ,200 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 e. Floc Removal (dry) CY 2, ,761 Annual Total: $ 491,953 Partial Floc Capture OPTION UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 374, ,992 b. Labor (20 hours/week) man-hours 1, ,200 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 e. Floc Removal (dry) CY 1, ,381 Annual Total: $ 471,572

64 4-34 TABLE CONTINUED ESTIMATED ANNUAL O&M COSTS FOR THE ALUM INFLOW TREATMENT OPTIONS Floc Discharge to Lake OPTION 53,250 kg TP/yr Removal Option UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 1,253, ,190,411 b. Labor (24 hours/week) man-hours 1, ,440 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 Annual Total: $ 1,326,851 Full Floc Capture OPTION UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 1,253, ,190,411 b. Labor (32 hours/week) man-hours 1, ,920 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 e. Floc Removal (dry) CY 6, ,303 Annual Total: $ 1,475,634 Partial Floc Capture OPTION UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 1,253, ,190,411 b. Labor (32 hours/week) man-hours 1, ,920 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 e. Floc Removal (dry) CY 3, ,15 Annual Total: $ 1,407,483

65 4-35 TABLE CONTINUED ESTIMATED ANNUAL O&M COSTS FOR THE ALUM INFLOW TREATMENT OPTIONS Floc Discharge to Lake OPTION 68,160 kg TP/yr Removal Option UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 1,508, ,433,319 b. Labor (24 hours/week) man-hours 1, ,440 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 Annual Total: $ 1,569,759 Full Floc Capture OPTION UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 1,508, ,433,319 b. Labor (32 hours/week) man-hours 1, ,920 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 e. Floc Removal (dry) CY 8, ,116 Annual Total: $ 1,746,355 Partial Floc Capture OPTION UNITS ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) a. Liquid Alum gallons 1,508, ,433,319 b. Labor (32 hours/week) man-hours 1, ,920 c. Utilities LS ,000 d. Repairs & Miscellaneous Items LS ,000 e. Floc Removal (dry) CY 4, ,058 Annual Total: $ 1,664,297

66 4-36 TABLE 4-16 CALCULATED PHOSPHORUS REMOVAL COSTS FOR THE 2,500 kg TP/yr REMOVAL OPTION A. Floc Discharge to Lake PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 2,500 kg/yr 50,000 kg $ 1,616,456 $ 133,618 $ 2,977, Removal Cost $ 59.50/kg B. Full Floc Capture PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 2,500 kg/yr 50,000 kg $ 1,776,474 $ 148,428 $ 3,793, Removal Cost $ 75.90/kg C. Partial Floc Capture PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 2,500 kg/yr 50,000 kg $ 1,538,005 $ 144,923 $ 3,507, Removal Cost $ 70.20/kg

67 4-37 A summary of 20-year present worth phosphorus removal costs for the 14,500 kg TP/yr removal option is given in Table The calculations summarized in this table were conducted in the same manner as previously described for the 2,500 kg TP/yr removal option. The present worth phosphorus removal cost for direct floc discharge to Barr Lake is $25.00/kg. If full floc capture is added, the phosphorus removal cost increases to $41.00/kg. Under partial floc capture conditions, the phosphorus removal cost is estimated to be approximately $33.70/kg. A summary of 20-year present worth phosphorus removal costs for the 53,250 kg TP/yr removal option is given in Table If the system provides for full floc discharge into Barr Lake, the phosphorus removal cost is $18.20/kg. If full capture of the generated floc is desired, then the phosphorus removal cost increases to $30.20/kg. The partial floc collection option results in a phosphorus removal cost of $24.60/kg, approximately mid-way between the direct floc discharge and full floc capture option. A summary of 20-year present worth phosphorus removal costs for the 68,160 kg TP/yr removal option is given in Table For the system which allows direct floc discharge to Barr Lake, the phosphorus removal cost is only $16.60/kg of phosphorus removed. Full floc capture increases the removal cost to $27.80/kg, with partial floc capture resulting in a phosphorus removal cost of $22.50/kg. Each of the phosphorus removal costs summarized in Tables 4-16 through 4-19 reflect extremely low phosphorus removal costs. Typical phosphorus removal costs for wet detention ponds range from approximately $500-1,000/kg, with substantially more elevated phosphorus removal costs for other removal systems Treatment System Operation A summary of operational procedures for the evaluated alum inflow treatment options is given on Table Operational procedures are provided for the direct floc discharge option, the full floc collection option, and the partial floc collection option. 4.5 Media Filtration The media filtration option consists of a vertical downflow filter which uses a specially designed media blend to absorb soluble nutrients from the incoming flow. A large number of media blends are currently available which have specific infiltration and adsorption characteristics. Typical components of a filtration media include sand, clay, tire crumb, peat, and other materials specifically designed to produce desired hydrologic and uptake characteristics. Removal processes in the media occur as a result of direct filtration of solid particulate matter from the water as well as adsorption of dissolved constituents onto the media particles. Aluminum-based media are commonly used for phosphorus removal, including residual sludges from water treatment processes. However, selection of the optimum media for this application would be based upon laboratory testing using water collected from the Burlington Ditch.

68 4-38 TABLE 4-17 CALCULATED PHOSPHORUS REMOVAL COSTS FOR THE 14,500 kg TP/yr REMOVAL OPTION A. Floc Discharge to Lake PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 14,500 kg/yr 290,000 kg $ 1,211,256 $ 443,392 $ 7,236, Removal Cost $ 25.00/kg B. Full Floc Capture PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 14,500 kg/yr 290,000 kg $ 5,212,658 $ 491,953 $ 11,898, Removal Cost $ 41.00/kg C. Partial Floc Capture PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 14,500 kg/yr 290,000 kg $ 3,369,980 $ 471,572 $ 9,778, Removal Cost $ 33.70/kg

69 4-39 TABLE 4-18 CALCULATED PHOSPHORUS REMOVAL COSTS FOR THE 53,250 kg TP/yr REMOVAL OPTION A. Floc Discharge to Lake PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 53,250 kg/yr 1,065,000 kg $ 1,334,856 $ 1,326,851 $ 19,366, Removal Cost $ 18.20/kg B. Full Floc Capture PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 53,250 kg/yr 1,065,000 kg $ 12,085,361 $ 1,475,634 $ 32,139, Removal Cost $ 30.20/kg C. Partial Floc Capture PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 53,250 kg/yr 1,065,000 kg $ 7,022,987 $ 1,407,483 $ 26,150, Removal Cost $ 24.60/kg

70 4-40 TABLE 4-19 CALCULATED PHOSPHORUS REMOVAL COSTS FOR THE 68,160 kg TP/yr REMOVAL OPTION A. Floc Discharge to Lake PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 68,160 kg/yr 1,363,200 kg $ 1,334,856 $ 1,569,759 $ 22,667, Removal Cost $ 16.60/kg B. Full Floc Capture PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 68,160 kg/yr 1,363,200 kg $ 14,205,839 $ 1,746,355 $ 37,938, Removal Cost $ 27.80/kg C. Partial Floc Capture PARAMETER 1. Load Reduction a. Annual Removal b. 20-year Removal 2. Costs a. Construction b. Annual O&M c. 20-year Present Worth TOTAL P 68,160 kg/yr 1,363,200 kg $ 8,097,689 $ 1,664,297 $ 37,715, Removal Cost $ 22.50/kg

71 4-41 TABLE 4-20 OPERATIONAL PROCEDURES FOR THE ALUM INFLOW TREATMENT OPTION Direct Floc Discharge Option a. Close existing bifurcation structure control gates into Barr Lake b. Open sluice gates to divert flow into off-line channel c. Flow created by head difference in ditch between areas upstream and downstream from bifurcation structure d. Flow measured in channel e. Obtain desired inflow rate by adjusting sluice gate(s) f. Alum added on flow-proportional basis g. Alum treated water containing floc diverted back into Burlington Ditch h. Floc is carried to Barr Lake and settles onto lake bottom Full Floc Collection Option a. Close existing bifurcation structure control gates into Barr Lake b. Open sluice gates in the inflow and outflow channels c. Flow created by head difference in ditch between areas upstream and downstream from bifurcation structure d. Flow measured in inflow culvert e. Obtain desired inflow rate by adjusting sluice gates f. Alum added on flow-proportional basis g. Alum treated water containing floc discharges into floc collection pond h. Floc settles onto bottom of settling pond i. Treated water discharges into Burlington Ditch downstream from bifurcation control structure j. Floc removed during dry season Partial Floc Collection Option a. Floc collection to occur primarily during high water levels b. Close existing bifurcation structure control gates into Barr Lake c. Open sluice gates in the inflow and outflow channels d. Flow created by head difference in ditch between areas upstream and downstream from bifurcation structure e. Obtain desired inflow rate by adjusting sluice gates f. Flow measured in inflow culvert g. Alum added on flow-proportional basis h. Alum treated water containing floc discharges into floc collection pond i. Floc settles onto bottom of settling pond j. Treated water discharges into Burlington Ditch downstream from bifurcation control structure k. Floc removed during dry season

72 4-42 Due to the large volumes of water which must be processed, the media filtration designs are based upon gravity flow rather than continuous pumping systems. Water from the Burlington Ditch would be introduced into the filtration bed area and allowed to infiltrate through the media where the adsorption and removal processes occur. The treated water is then collected in an underdrain system and discharged into the receiving waterbody Basic Media Filtration System Components A cross-section for a typical media filtration system is given on Figure For the system to operate by gravity, the top of the filtration media must be equal to or less than the bottom of the Burlington Ditch channel so that water would be capable of flowing into the filtration system even during periods of low water conditions. Water would be diverted from the Burlington Ditch into an off-line inflow channel with multiple inflow pipes used to connect the inflow channel to the filtration system. A distribution flume would be used to distribute the water evenly over the media area. Water from the ditch would flow into the filter media area, migrating downward through the media, and be collected in a perforated underdrain system. The perforated underdrain system would be connected to a central collection piping system which would collect and discharge the treated water through a series of outflow pipes into the outflow canal for return to the Burlington Ditch. To operate by gravity, the system would be constructed with the inflow upstream of the bifurcation structure and the discharge downstream of the bifurcation structure so that the head difference created by the bifurcation structure could be used to force the water through the system. A cross-section of the media portion of the filter system is given on Figure The filtration media will have a depth of approximately 2 ft which is placed on top of a 6-inch layer of sand. The sand layer separates the filtration media from the underdrain gravel system and allows the media to be removed without disturbing the underdrains. The underdrain system consists of a gravel layer approximately 1 ft in height which is crisscrossed with a series of 6- inch perforated underdrains that collect the filtered water and transmit to the point of discharge in the outflow canal. A summary of media filtration system design parameters for the evaluated load reduction options is given on Table Maximum desired treated inflow rates are provided for each of the three load reduction options based upon the summary of treatment design criteria provided in Table 4-3. Maximum treated inflow rates range from cfs for the three options, with annual treated volumes ranging from 2,185-41,261 ac-ft/yr. The filtration system is designed based upon an assumed hydraulic rate of 15 inches/hour. This is a typical infiltration rate for media filtration systems and is a function of the selected media components for a given system. This analysis assumes a media adsorption capacity of 3 g P/kg filter media which is a conservative estimate. The density of the filtration media is assumed to be approximately 55 lbs/ft 3.

73 4-43 Inflow channel from Burlington Ditch Rip rap 24 HDPE inflow piping Filter media 6 sand 12 gravel Inflow distribution channel Liner 6 perforated underdrain Outflow channel to Barr Lake Liner 6 perforated underdrain 12 HDPE collection pipe 12 HDPE outflow piping Figure Cross-Section Through Media Filtration System. 2 ft. Filtration Media 6 6 in. Sand 1 ft. Gravel 6 Perforated underdrain Liner Figure Cross-Section of Media Portion of Filter System.

74 4-44 TABLE 4-21 MEDIA FILTRATION SYSTEM DESIGN PARAMETERS FOR THE EVALUATED LOAD REDUCTION OPTIONS PARAMETER UNITS LOAD REDUCTION OPTION 2,500 kg TP/yr 14,500 kg TP/yr 53,250 kg TP/yr Annual Required TP Removal kg 2,500 14,500 53,250 Maximum Treated Inflow cfs Average Daily Treated Inflow cfs ac-ft/day Annual Treated Volume ac-ft/yr 2,185 12,580 41,261 Filter Hydraulic Loading Rate in/hr Media Adsorption Capacity g P/kg Design Filter Lifespan years Media Density lbs/ft Media Requirements kg 1,000,000 yd 3 2,970 5,800,000 17,224 21,300,000 63,255 Media Depth inches Filter Area acres Media Cost $/yd $ 534,545 3,100,364 11,385,818 Estimates of media requirements were generated by dividing the annual required TP removal by the media adsorption capacity. The filtration media is designed for a 2-year useable life so the required annual media volume is doubled to obtain media requirements for a 2-year cycle. The required media volumes for the three evaluated load reduction options range from 2,970-63,255 yd 3. The resulting filtration system area requirements range from acres, depending upon the load reduction option. At an assumed media cost of $180/yd 3, the media cost (based upon a 2-year life cycle) ranges from approximately $534,545-11,385,818 depending upon the load reduction option Conceptual Designs and Costs A conceptual schematic of an off-line media filtration system for the 2,500 kg TP/yr treatment option is given on Figure Water from the Burlington Ditch would be diverted through a 36-inch HDPE equipped with a sluice gate to regulate inflow rates. The ditch water would discharge into an inflow canal and be transferred into the filter bed through a distribution channel. The water would then filter through the media, be collected in a series of 6-inch perforated underdrains, and discharged to the outflow canal for return back to the Burlington Ditch. A sluice gate would also be placed on the discharge side of the piping connected to the Burlington Ditch so that the off-line filtration system could be isolated from the ditch if desired.

75 HDPE outflow piping 6 perforated underdrain Outflow Canal Rip-rap 36 HDPE Half pipe Distribution channel Filter Bed Pedestrian bridge Inflow Canal 36 HDPE with sluice gate 36 HDPE with sluice gate Figure Conceptual Schematic of an Off-line Media Filtration System for the 2,500 kg TP/yr Treatment Option. A conceptual cost estimate for media filtration with the 2,500 kg TP/yr removal option is given in Table As indicated on Table 4-21, the required filter area for this option is 0.9 acres. An additional 25% allowance is included for maintenance areas and roadways, resulting in a land requirement of approximately 1.25 acres. Costs are provided for excavation of the filter basin, inflow and outflow canals, rip-rap, piping, motorized sluice gates, crushed gravel, sand, filter media, and the bottom liner. The overall estimated construction cost for media filtration with the 2,500 kg TP/yr removal option is approximately $2,234,881, including a 20% contingency. A cost estimate is also provided on Table 4-22 for removal and replacement of the filtration media at two year intervals. Filter media removal is assumed at a unit price of $25/yd 3, with media replacement cost assumed to be $200/yd 3 which includes the cost of the media plus installation. The estimated bi-annual media replacement cost is approximately $801,818. A conceptual schematic of an off-line media filtration system for the 14,500 kg TP/yr treatment option is given on Figure As indicated Table 4-20, this option requires a filter area of approximately 5.3 acres. An additional 25% land allowance is included for maintenance areas and roadways.

76 4-46 TABLE 4-22 CONCEPTUAL COST ESTIMATE FOR MEDIA FILTRATION WITH THE 2,500 kg TP/yr REMOVAL OPTION NO. ITEM DESCRIPTION UNITS QUANTITY UNIT PRICE ($) COST ($) 1. Land ACRES ,000 31, Site Work LS 1 25,000 25, Excavation a. Filter Basin b. Inflow Canal c. Outflow Canal CY CY CY 37, , ,720 15,620 40, Rip-Rap CY , inch HDPE Culvert LF , inch Sluice Gate (motorized) EA 2 8,000 16, inch Perforated Underdrain LF 1, , inch HDPE piping LF , inch HDPE piping LF , Crushed Gravel CY 1, , Sand CY , Filter Media CY 2, , Liner SY 4, ,400 Sub-Total: $ 1,862,401 Contingency (20%) 372,480 TOTAL: $ 2,234,881 Media Replacement (2-year Intervals) NO. ITEM DESCRIPTION UNITS QUANTITY UNIT PRICE ($) COST ($) 1. Filter Media Removal CY 2, , Filter Media Replacement CY 2, ,939 Sub-Total: $ 668,182 Contingency (20%) 133,636 TOTAL: $ 801,818

77 HDPE Half pipe Distribution channel 6 perforated underdrain 12 HDPE outflow piping Rip-rap Inflow Canal Filter Bed Outflow Canal Pedestrian bridge 4 x 6 CBC with sluice gate 4 x 6 CBC with sluice gate Figure Conceptual Schematic of an Off-line Media Filtration System for the 14,500 kg TP/yr Treatment Option. A conceptual cost estimate for media filtration with the 14,500 kg TP/yr removal option is given on Table This cost estimate was generated in the same manner as previously described for the 2,500 kg TP/yr option. The estimated construction cost for the 14,500 kg TP/yr removal option is $11,097,264, including a 20% contingency. The bi-annual filter media replacement for this option is $4,650,545. A conceptual schematic of the media filtration system for the 53,250 kg TP/yr removal option was not developed since the required filter area of 19.6 acres substantially exceeds the land currently available adjacent to the bifurcation structure. Since the specific parcels which may be available will impact the configuration and design of the system, a hypothetical design is assumed by multiplying the calculated quantities for the 14,500 kg TP/yr option times the ratio of the pond areas for the 53,250 kg TP/yr and 14,500 kg TP/yr options (19.6 acres 5.3 acres). A summary of conceptual cost estimates for media filtration with the 53,250 kg TP/yr option is given on Table The estimated construction cost for the media filtration system is approximately $40,454,382, with a bi-annual media replacement cost of approximately $15,560,618.

78 4-48 TABLE 4-23 CONCEPTUAL COST ESTIMATE FOR MEDIA FILTRATION WITH THE 14,500 kg TP/yr REMOVAL OPTION NO. ITEM DESCRIPTION UNITS QUANTITY UNIT PRICE ($) COST ($) 1. Land ACRES , , Site Work LS 1 65,000 65, Excavation a. Filter Basin b. Inflow Canal c. Outflow Canal CY CY CY 170,964 3,661 3, ,419,280 73,220 60, Rip-Rap CY 1, , ft x 6-ft Concrete Box Culvert LF , ft x 6-ft Sluice Gate (motorized) EA 4 16,000 64, inch Perforated Underdrain LF 12, , inch HDPE piping LF 1, , inch HDPE piping LF , Crushed Gravel CY 8, , Sand CY 4, , Filter Media CY 17, ,100, Liner SY 25, ,520 Sub-Total: $ 9,247,720 Contingency (20%) 1,849,544 TOTAL: $ 11,097,264 Media Replacement (2-year Intervals) NO. ITEM DESCRIPTION UNITS QUANTITY UNIT PRICE ($) COST ($) 1. Filter Media Removal CY 17, , Filter Media Replacement CY 17, ,444,848 Sub-Total: $ 3,875,455 Contingency (20%) 775,091 TOTAL: $ 4,650,545

79 4-49 TABLE 4-24 CONCEPTUAL COST ESTIMATE FOR MEDIA FILTRATION WITH THE 53,250 kg TP/yr REMOVAL OPTION NO. ITEM DESCRIPTION UNITS QUANTITY UNIT PRICE ($) COST ($) 1. Land ACRES , , Site Work LS 1 90,000 90, Excavation a. Filter Basin b. Inflow Canal c. Outflow Canal CY CY CY 632,244 13,539 11, ,644, , , Rip-Rap CY 5, , ft x 6-ft Concrete Box Culvert LF , ft x 6-ft Sluice Gate (motorized) EA 8 16, , inch Perforated Underdrain LF 47, ,356, inch HDPE piping LF 5, , inch HDPE piping LF 2, , Crushed Gravel CY 31, ,162, Sand CY 15, , Filter Media CY 63, ,385, Liner SY 94, ,640 Sub-Total: $ 33,711,985 Contingency (20%) 6,742,397 TOTAL: $ 40,454,382 Media Replacement (2-year Intervals) NO. ITEM DESCRIPTION UNITS QUANTITY UNIT PRICE ($) COST ($) 1. Filter Media Removal CY 63, ,581, Filter Media Replacement CY 63, ,385,818 Sub-Total: $ 12,967,182 Contingency (20%) 2,593,436 TOTAL: $ 15,560,618

80 Annual O&M Costs A summary of estimated annual O&M costs for the media filtration options is given on Table Costs are included for operation of the system, with 10 hours/week assumed for the 2,500 kg TP/yr option, 20 hours/week assumed for the 14,500 kg TP/yr option, and 30 hours/week assumed for the 53,250 kg TP/yr option. Annual costs for repairs and miscellaneous items are also included, with increasing allocations for increasing system size. Utilities are assumed at a constant rate of $1,000/month. The estimated annual O&M costs for the media filtration options range from $52, ,800/yr, excluding media replacement costs Present Worth and Phosphorus Removal Costs A summary of calculated phosphorus removal costs for the media filtration options is given on Table Estimates of the annual and 20-year phosphorus load reductions are provided for each of the three options. Construction costs for the options are based upon information provided in Tables 4-22 through 4-24, with annual O&M costs obtained from Table The 20-year present worth cost is calculated by adding the construction cost to 20 years of annual O&M costs (based on a multiplier), and bi-annual media replacement costs with appropriate factors to convert future costs into present worth. The resulting 20-year present worth costs range from $7,925, ,835,696 for the three phosphorus removal options. Estimated phosphorus removal costs for this option are substantially greater than obtained with the alum treatment option, ranging from $159/kg for the 2,500 kg TP/yr option to $130/kg for the 53,250 kg TP/yr option. However, even though the phosphorus removal costs are substantially larger than the alum treatment removal costs, the media filtration costs are still substantially less than typical phosphorus removal costs associated with typical treatment systems Treatment System Operation A summary of operational procedures for the media filtration treatment options is given on Table The operational procedures summarized on Table 4-27 apply to each of the evaluated media filtration options.

81 4-51 TABLE 4-25 ESTIMATED ANNUAL O&M COSTS FOR THE MEDIA FILTRATION OPTIONS OPTION UNITS 2,500 kg TP/yr Option ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) Labor (10 hours/week) man-hours ,600 Repairs and Miscellaneous Items LS ,000 Utilities $/month 12 1,000 12,000 Annual Total: $ 52,600 OPTION UNITS 14,500 kg TP/yr Option ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) Labor (20 hours/week) man-hours 1, ,200 Repairs and Miscellaneous Items LS ,000 Utilities $/month 12 1,000 12,000 Annual Total: $ 93,200 OPTION UNITS 53,250 kg TP/yr Option ANNUAL QUANTITY UNIT COST ($) ANNUAL COST ($) Labor (30 hours/week) man-hours 1, ,800 Repairs and Miscellaneous Items LS ,000 Utilities $/month 12 1,000 12,000 Annual Total: $ 133,800

82 4-52 TABLE 4-26 CALCULATED PRESENT WORTH AND PHOSPHORUS REMOVAL COSTS FOR THE MEDIA FILTRATION OPTION PARAMETER Load Reduction a. Annual b. 20-Year Costs a. Construction b. Annual O&M c. 20-Year Present Worth 2,500 kg TP/yr Option UNITS kg/yr kg $ $ $ TOTAL PHOSPHORUS 2,500 50,000 2,234,881 52,600 7,925,478 Removal Cost $/kg PARAMETER Load Reduction a. Annual b. 20-Year Costs a. Construction b. Annual O&M c. 20-Year Present Worth 14,500 kg TP/yr Option UNITS kg/yr kg $ $ $ TOTAL PHOSPHORUS 14, ,000 11,097,264 93,200 41,223,277 Removal Cost $/kg PARAMETER Load Reduction a. Annual b. 20-Year Costs a. Construction b. Annual O&M c. 20-Year Present Worth 53,250 kg TP/yr Option UNITS kg/yr kg $ $ $ TOTAL PHOSPHORUS 53,250 1,065,000 40,454, , ,835,696 Removal Cost $/kg 130.4

83 4-53 TABLE 4-27 OPERATIONAL PROCEDURES FOR THE MEDIA FILTRATION OPTION a. Close existing bifurcation structure control gates into Barr Lake b. Open sluice gates in inflow and outfall channels c. Flow measured in inflow culvert d. Adjust sluice gates to obtain the desired inflow e. Ditch water diverted onto filtration media f. Treated water discharges into Burlington Ditch downstream from water control structure g. Filter dries out during dry season

84 SECTION 5 SUMMARY Load reduction options were evaluated for alum treatment and media filtration systems to reduce phosphorus loadings to Barr Lake for load reduction scenarios specified by BMW. Four separate options were evaluated for both alum treatment and media filtration. The first option (Option 1), which specifies a required annual total phosphorus reduction of 2,500 kg/yr, is based upon a 75% reduction of loadings from Cherry Creek, Bear Creek, and Chatfield Reservoirs which have a combined total phosphorus input of approximately 3,000 kg/yr. Option 2 includes a 75% reduction of loadings from Cherry Creek, Bear Creek, and Chatfield Reservoirs plus point source loadings discharging to Barr Lake. This option has a required annual total phosphorus load reduction of approximately 14,500 kg/yr. Option 3 provides for a 75% removal of the total annual phosphorus loading to Barr Lake from the Burlington Canal, equivalent to an annual total phosphorus reduction of 53,250 kg/yr. The final option (Option 4) provides for a removal of 96% of the total annual phosphorus loadings to Bear Lake which is equivalent to the total TMDL required load reductions to the lake. The annual total phosphorus load reduction for this option is 68,160 kg/yr. Each of these four load reduction scenarios were evaluated for both the alum treatment and media filtration options. Conceptual designs were developed for each of the evaluated load reduction options, including estimates of construction, O&M, and phosphorus removal costs. A summary of estimated construction, O&M, and phosphorus mass removal costs for the alum inflow treatment options is given on Table 5-1. Estimated construction costs for the alum treatment options vary substantially depending upon the required annual phosphorus mass load reduction and method of floc collection. Calculated phosphorus removal costs are extremely low for each of the evaluated options compared with phosphorus removal costs commonly associated with stormwater BMPs. For the 68,160 kg TP/yr removal option, phosphorus removal costs ranged from $ /kg depending upon method of floc capture. These costs are extremely low in value and substantially lower than any previously reported phosphorus stormwater removal costs. A summary of estimated construction, O&M, and phosphorus removal costs for the media filtration option is given on Table 5-2. Estimated construction costs for this option range from $2,234,881 for the 2,500 kg TP/yr option to $40,454,382 for the 53,250 kg TP/yr option. The media filtration is not capable of achieving the 68,160 kg TP/yr option. Calculated mass removal costs for the filtration option are substantially higher than estimated for the alum inflow treatment option, although lower in value than typical BMP removal costs. 5-1

85 5-2 TABLE 5-1 SUMMARY OF CONSTRUCTION, O&M, AND TOTAL PHOSPHORUS MASS REMOVAL COSTS FOR THE ALUM INFLOW TREATMENT OPTIONS OPTION 1. 2,500 kg TP/year a. Floc Discharge to Lake b. Full Floc Capture c. Partial Floc Capture 2. 14,500 kg TP/year a. Floc Discharge to Lake b. Full Floc Capture c. Partial Floc Capture 3. 53,250 kg TP/year a. Floc Discharge to Lake b. Full Floc Capture c. Partial Floc Capture 4. 68,160 kg TP/year a. Floc Discharge to Lake b. Full Floc Capture c. Partial Floc Capture CONSTRUCTION COST ($) 1,161,456 1,776,474 1,538,005 1,211,256 5,198,258 3,369,980 1,334,856 12,085,361 7,022,987 1,334,856 14,205,839 8,097,689 ANNUAL O&M COST ($) 133, , , , , ,572 1,270,611 1,372,101 1,321,356 1,513,519 1,615,009 1,564,264 MASS TP REMOVAL COST 1 ($/kg) year present worth cost, i = 4% TABLE 5-2 OPTION SUMMARY OF CONSTRUCTION, O&M, AND TOTAL PHOSPHORUS MASS REMOVAL COSTS FOR THE MEDIA FILTRATION OPTION CONSTRUCTION COST ($) ANNUAL O&M COST ($) BI-ANNUAL MEDIA REPLACEMENT ($) MASS TP REMOVAL COST 1 ($/kg) 1. 2,500 kg TP/year 2,234,881 52, , ,500 kg TP/year 11,097,264 93,200 4,650, ,250 kg TP/year 40,454, ,800 15,560, year present worth cost, i = 4%

86 APPENDICES

87 APPENDIX A RESULTS OF LABORATORY JAR TESTING CONDUCTED ON BURLINGTON DITCH INFLOWS TO BARR LAKE

88 Sample Description Date Collected Results of Jar Test Samples Conducted on Barr Lake Inflow Collected on January 28, 2014 ph (s.u.) Alkalinity Conductivity NH 3 NO X Diss. Org. N Part. N Total N SRP Diss. Org. P Part. P Total P Color Diss. Al Initial 1 Min. 1 Hour 24 Hours (mg/l) (µmho/cm) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (Pt-Co) (µg/l) Raw 1/28/ , , , mg/l 1/28/ , , , mg/l 1/28/ , , , mg/l 1/28/ , , , mg/l 1/28/ , , , mg/l 1/28/ , , , mg/l 1/28/ , , , Change in Phosphorus Species Sample Change in Concentration (%) Description SRP Diss. Org. P Part. P Total P Color Raw mg/l mg/l mg/l mg/l mg/l mg/l Sample Description Date Collected Results of Jar Test Samples Conducted on Barr Lake Inflow Collected on March 14, 2014 ph (s.u.) Alkalinity Conductivity NH 3 NO X Diss. Org. N Part. N Total N SRP Diss. Org. P Part. P Total P Color Diss. Al Initial 1 Min. 1 Hour 24 Hours (mg/l) (µmho/cm) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (Pt-Co) (µg/l) Raw 3/14/ , , , mg/l 3/15/ , , , mg/l 3/15/ , , , mg/l 3/15/ , , , mg/l 3/15/ , , , mg/l 3/15/ , , , mg/l 3/15/ , , , Change in Phosphorus Species Sample Change in Concentration (%) Description SRP Diss. Org. P Part. P Total P Color Raw mg/l mg/l mg/l mg/l mg/l mg/l

89 Sample Description Date Collected Results of Jar Test Samples Conducted on Barr Lake Inflow Collected on April 15, 2014 ph (s.u.) Alkalinity Conductivity NH 3 NO X Diss. Org. N Part. N Total N SRP Diss. Org. P Part. P Total P Color Diss. Al Initial 1 Min. 1 Hour 24 Hours (mg/l) (µmho/cm) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (Pt-Co) (µg/l) Raw 4/15/ ,001 ` 2, , mg/l 4/15/ , , , mg/l 4/15/ , , , mg/l 4/15/ , , , mg/l 4/15/ , , , mg/l 4/15/ , , , mg/l 4/15/ , , , Change in Phosphorus Species Sample Change in Concentration (%) Description SRP Diss. Org. P Part. P Total P Color Raw mg/l mg/l mg/l mg/l mg/l mg/l Mean Characteristics of Jar Tests Conducted on Barr Lake Inflow ph (s.u.) Alkalinity Conductivity NH 3 NO X Diss. Org. N Part. N Total N SRP Diss. Org. P Part. P Total P Color Diss. Al Initial 1 Min. 1 Hour 24 Hours (mg/l) (µmho/cm) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (µg/l) (Pt-Co) (µg/l) Raw , , , Sample Description 2.0 mg/l , , , mg/l , , , mg/l , , , mg/l , , , mg/l , , , mg/l , , , Change in Phosphorus Species Sample Change in Concentration (%) Description SRP Diss. Org. P Part. P Total P Color Raw mg/l mg/l mg/l mg/l mg/l mg/l

90 APPENDIX B SECTION 2 OF THE TECHNICAL REPORT TITLED TECHNICAL ASSISTANCE FOR THE NORTHERN EVERGLADES CHEMICAL TREATMENT PILOT PROJECT (July 6, 2009)

91 TECHNICAL ASSISTANCE FOR THE NORTHERN EVERGLADES CHEMICAL TREATMENT PILOT PROJECT By: Del Bottcher Tom DeBusk Harvey Harper Steve Iwinski George O Connor, and Marty Wanielista PROJECT OFFICER James Laing SFWMD PROJECT ID#: PS July 6, 2009

92 Final Report Chemical Treatment ABSTRACT Chemical treatment of stormwater discharges for the purpose of phosphorus removal is under consideration for use within the South Florida Water Management District (SFWMD). There exists significant literature on the subject but no one document which summarizes information useful for decision making within the SFWMD. A significant investment in time, effort and money may be allocated to chemical treatment, thus the purpose of this report is to present a review of existing information and an evaluation of potential uses of chemical treatment to reduce phosphorus loading within the Kissimmee, Okeechobee, and Everglades (KOE) areas and at varying scales. An extensive literature review is complete. Approximately 80% of the literature cited was in a form that could be copied without copyright violation and reproduced in electronic form. Thus an electronic copy for most of the literature and is available and entitled Depository for Support Literature. Other literature not cited in the report is also included in the electronic copy and together all the literature provide an in-depth review. In reviewing available information, the authors are guided by the following implementation issues. 1. What P concentrations and/or species will respond to chemical treatment cost effectively? 2. What volume or flow rate is logistically feasible for treatment? 3. Where in the KOE planned features can chemical treatment be applied? 4. What water quality parameters affect chemical treatment p-reduction efficiency? Do we have sufficient existing data or is additional data required? 5. What water quality standards must be met for chemically treated discharges to various receiving waters? 6. What is the best aerial economy of scale for treatment system implementation (parcel, sub-basin, STA, reservoir)? 7. Can the chemical treatment be permitted? 8. What are the monitoring requirements of planned solutions? 9. What are the cost-benefits of planned solutions? 10. What factors affect settling and residuals management? 11. What are cost effective options for residual management? 12. What chemicals and treatment configurations should be further evaluated? The existing information shows the use of chemicals to control phosphorus that discharge from watersheds is well established and may be a viable and cost effective option for consideration by decision makers in the SFWMD. The practical application of the technology requires an understanding of flow attenuation, chemical dosing equipment, sludge handling, and local cost 7

93 Final Report Chemical Treatment considerations. Long term cost to include initial and operating cost investments favor larger areas because of the reduced variability of flow and concentration. Nevertheless, chemical treatment may be cost effective and reliable for most watershed sizes provided the operation can be justified and the terms of a permit can be met. Monitoring of the effluent should not be a problem. Cost comparisons can be made as chemical treatment is cost effective for most watershed sizes. However if the treatment train is land intensive, land cost may be significant and may be included in comparisons. Because of the many combinations of chemical treatment options, a cost comparison among the various methods is not reasonable; nevertheless comparisons are made with the knowledge that they are not always completely comparable. A majority of the applications for flows from a concentrated discharge have used aluminum sulfate (alum) compounds. Enhanced removal when using a chemical metal salt may be achieved with an anionic polymer. Dosage and toxicity tests should be conducted before design and construction to determine dosages and effectiveness. The literature supports the method of dosage determination as a standard procedure well known by many professionals. The use of water treatment residuals (WRT) is an option for areal treatment, and can also be used for concentrated source treatment, but availability of WRT may be an issue. Laboratory testing should be done to determine optimal dosage. It is also known and documented that chemical treatment can be used as part of a combination of treatment methods. Detention facilities, edge of farm (EOF) and Hybrid Wetland Treatment Technologies (HWTT) are available for consideration. The HWTT system concept does reduce some of the problems associated with residual management as the residuals can be reused in the system. Residual management must be a consideration for the residuals and is part of cost effective calculations. Lastly, permitting of the chemical treatment methods is a function of the regulatory agencies and with the data presented in this report together with the comprehensive associated literature list, the authors believe that the system can be permitted. 8

94 Final Report Chemical Treatment CHAPTER II ALUMINUM, IRON, AND CALCIUM SALTS INTRODUCTION Coagulation and clarification of water using metal salts has been practiced since at least Roman times to reduce turbidity and improve the appearance of drinking water and surface water. The predominant chemical agent used in these processes has been aluminum sulfate [Al 2 (SO 4 ) 3 ], commonly referred to as alum. Lime [Ca(OH) 2 ] has also been used, either alone or in combination with alum as well as with iron salts, such as ferric sulfate [Fe 2 (SO 4 ) 3 ] or ferric chloride (FeCl 3 ). Alum was used by the ancient Romans beginning around 2000 BC as a coagulant which was mixed with lime to make bitter water potable. Beginning in the mid-1700s, muddy water in England was treated with alum, followed by flocculation and filtration of the supernatant, to improve the quality of drinking water. Large-scale coagulation of municipal water supplies originated in Baltom, England in 1881 (Baker, 1981). The first scientific investigation into the use of alum for coagulation in the United States was conducted by Rutgers University in They concluded alum was useful in clarifying turbid water without impairment to taste or physiological properties. During , a series of experiments was conducted on turbid water collected from the Ohio River. A variety of compounds were tested, including alum, potash, and lime, with alum found to be the most suitable. These experiments eventually led to the widespread use of alum coagulation in the United States. Concurrent research was also conducted on the use of iron compounds, such as ferrous sulfate and ferric chloride, which were found to be reasonably effective in certain situations. However, alum remains the most widely used coagulant today. In 1970, Jernelov was apparently the first to use alum to remove phosphorus from the water column of a lake in a whole-lake alum application conducted as part of a lake restoration project on Lake Langsjon in Sweden. The first U.S. lake to be treated with a whole-lake alum application was Horseshoe Lake in Wisconsin which received a surface application of 2.6 mg Al/liter in May Twelve years later, phosphorus concentrations were still below the pretreatment level (Garrison and Knauer, 1984). In 1985, a lake restoration project was initiated at Lake Ella, a shallow, 13.3 ac hypereutrophic lake in Tallahassee, Florida, which receives untreated stormwater runoff from approximately 163 ac of highly impervious urban watershed areas. Initially, conventional stormwater treatment technologies, such as retention basins, exfiltration trenches and filter systems, were considered for reducing available stormwater loadings to Lake Ella in an effort to improve water quality within the lake. Since there was no available land surrounding Lake Ella that could be used for construction of traditional stormwater management facilities, and the cost of purchasing homes and businesses to acquire land for construction of these facilities was cost-prohibitive, alternate 15

95 Final Report Chemical Treatment stormwater treatment methods were considered. Chemical treatment of stormwater runoff was evaluated using various chemical coagulants, including alum, ferric salts and polymers. Alum consistently provided the highest removal efficiencies and produced the most stable end product. In view of successful jar test results on runoff samples collected from the Lake Ella watershed, the design of a prototype alum injection stormwater system was completed. Construction of the Lake Ella alum stormwater treatment system was completed in January 1987, resulting in a rapid and significant improvement in water quality. OVERVIEW OF TECHNOLOGY Characteristics of Common Coagulants A number of inorganic salts of calcium, iron, and aluminum are sold commercially for coagulation purposes. A summary of properties of common coagulants is given in Table 2-1. Within the United States, alum is used extensively for clarification of drinking water originating from surface water sources. Lime is commonly used for treatment of drinking water which originates as a groundwater source. Iron compounds are used predominantly in treatment of domestic and industrial wastewaters. Both aluminum and iron compounds are used for phosphorus removal in a variety of processes. TABLE 2-1 PROPERTIES OF COMMON COAGULANTS COMMON NAME FORMULA EQUIVALENT WEIGHT ph OF 1% SOLUTION AVAILABLE FORMS Alum Al 2 (SO 4 ) 3 p 14H 2 O Lump: 17% Al 2 O 3 Liquid: 4.4% Al Lime Ca(OH) Lump: Powder: Slurry: As CaO 93-95% 15-50% Ferric Chloride FeCl 3 p 6H 2 O Lump: Liquid: 20% Fe 10-45% FeCl 3 Ferric Sulfate Fe 2 (SO 4 ) 3 p 3H 2 O Granular: 18.5% Fe Copperas FeSO 4 p 7H 2 O Granular: 20% Fe Sodium Aluminate Na 2 Al 2 O Liquid: 20-25% Al 2 O % Al Aluminum Chloride AlCl 3 44 < 1 Liquid: 15-30% AlCl 3 Polyaluminum Hydroxychloride PACl Varies Varies Liquid: 3-13% Al 16

96 Final Report Chemical Treatment Alum is produced by dissolving aluminum ore in sulfuric acid and water. The most common aluminum sources used for production of alum are chemical grade bauxite, high aluminum clays, and aluminum trihydrate. Bauxite and bauxitic clays are used to produce the standard grade alum most commonly used for coagulation. The purity of alum will vary with aluminum and acid sources used in the production process. However, bauxite and bauxitic clays are low in metal contaminants, and alum solutions are typically low in most heavy metals. Aluminum chloride is generated in a similar manner by dissolving aluminum ore in hydrochloric acid. Polyaluminum hydroxychloride (PACl) consists of a variety of products which vary in both physical and chemical characteristics. Many PACl compounds contain supplemental hydroxide (OH) - ions which cause lower ph depression and alkalinity impacts during coagulation processes. The manufacturing of PACl in North America is commonly done with very pure raw materials. Thus, heavy metal impurities in PACl are often less than the cleanest standard alum. PACl is often a good choice for coagulation processes where ph depression is a significant concern. However, PACl is substantially more expensive than alum, and distributors are limited. Ferric sulfate has been used in the water treatment industry since the late 1800s. Ferric sulfate solutions can be either manufactured or reprocessed from waste streams generated in iron mills, foundries, and pickeling operations. One of the most common methods for manufacturing ferric sulfate is to dissolve iron ores or scrap iron in sulfuric acid. As a result, ferric sulfate is often highly variable in terms of its chemical composition and contaminant levels. Ferric chloride (FeCl 3 ) is the most widely used iron salt in North America, and is second only to alum for use in chemical coagulation. Ferric chloride is produced in a manner similar to ferric sulfate, where iron ore is dissolved in hydrochloric acid. As a result, heavy metals are common contaminants. Strict control of chemical characteristics of ferric chloride is necessary when using this compound in treating surface or drinking waters. Sodium aluminate is an alkaline form of alum which is formed by dissolving aluminum ore in sodium hydroxide. Sodium aluminate is a good choice for treatment of acidic waters since the excess alkalinity will provide ph neutralization. Sodium aluminate contains approximately three times as much aluminum by weight as alum, and as a result, must be used and dosed carefully to avoid overdose and undesirable increases in ph. The chemical impurities in sodium aluminate are similar to the level of impurities found in high grade alum. Unlike solutions of aluminum or iron which consist of dissolved ions in solution, lime is typically supplied as a slurry of calcium hydroxide solids in water. Since the product consists of a slurry of solids, it must be continuously agitated to prevent the solids from settling onto the bottom of the storage tank. Lime slurry is used in a wide variety of applications which include ph adjustment, metals precipitation, lime softening, coagulation, and sludge stabilization. Lime slurries have been used on a limited basis for removal of phosphorus in surface waters. However, lime precipitation typically occurs at a ph range of approximately which then requires ph neutralization as a second step. The lime precipitate must be separated from the treated water prior to ph neutralization to avoid dissolution of the lime precipitate and release of undesirable compounds as the ph is lowered. In view of the additional steps and equipment required for storage and distribution of lime, and the subsequent sludge separation and ph neutralization processes, lime is seldom used for coagulation processes designed to remove phosphorus. 17

97 Final Report Chemical Treatment A summary of typical analyses of common inorganic coagulants is given in Table 2-2. Both PACl and alum contain extremely low levels for virtually all of the heavy metals summarized in Table 2-2. In contrast, substantially higher metal concentrations are commonly observed in both ferric sulfate and ferric chloride due to the nature of the raw materials used to generate the products. Ferric chloride is often highly contaminated with manganese, titanium, vanadium, zinc, and chromium compared with PACl and alum. TABLE 2-2 TYPICAL ANALYSES OF INORGANIC COAGULANTS ELEMENT METAL CONCENTRATION (ppm) PACl Alum Fe 2 (SO 4 ) 3 FeCl 3 Silver < 0.4 < Barium < Cadmium < 0.05 < Cobalt < Chromium Copper < Manganese Nickel Titanium Vanadium Zinc Lead < 1 < Arsenic < 1 < Mercury < < SOURCE: WATER/Engineering & Management (Feb. 1998) Process Chemistry When aluminum sulfate is added to water, aluminum hydrous oxides are precipitated according to the following stoichiometric coagulation reaction: Al 2 (SO 4 ) 3 p 18H 2 O + 3Ca(HCO 3 ) 2 2Al(OH) 3(s) + 3CaSO 4 + 6CO 2(g) + 18H 2 O In this reaction, calcium carbonate is used to represent the alkalinity needed to form Al(OH) 3(s). According to this relationship, 1 mg/l of alum requires 0.45 mg/l of alkalinity as CaCO 3 and releases 0.9 mg/l of CO 2(g) as CaCO 3. The alum coagulation reaction is frequently abbreviated to 18

98 Final Report Chemical Treatment include just significant products and reactants. The addition of alum to water results in the production of chemical precipitates which remove pollutants by two primary mechanisms. Removal of suspended solids, algae, phosphorus, heavy metals and bacteria occurs primarily by enmeshment and adsorption onto aluminum hydroxide precipitate according to the following net reaction: Al H 2 O Al(OH) 3(s) + 3H 3 O + Removal of additional dissolved phosphorus occurs as a result of direct formation of AlPO 4 by: Al +3 + HnPO 4 n-3 AlPO 4(s) + nh + The aluminum hydroxide precipitate, Al(OH) 3, is a gelatinous floc which attracts and adsorbs colloidal particles onto the growing floc, thus clarifying the water. Phosphorus removal or entrapment can occur by several mechanisms, depending on the solution ph. Inorganic phosphorus is also effectively removed by adsorption to the Al(OH) 3 floc. Removal of particulate phosphorus is most effective in the ph range of 6-8 where maximum floc occurs (Cooke and Kennedy, 1981). At higher ph values, OH - begins to compete with phosphate ions for aluminum ions, and aluminum hydroxide-phosphate complexes begin to form. At lower ph values and higher inorganic phosphorus concentrations, the formation of aluminum phosphate (AlPO 4 ) is favored. The chemical stoichiometric reaction for coagulation with iron compounds is similar to the reactions previously provided for aluminum. A typical coagulation reaction involving ferric sulfate can be written as: Fe 2 (SO 4 ) 3 p 18H 2 O + 3Ca(HCO 3 ) 2 2Fe(OH) 3(s) + 3CaSO 4 + 6CO 2(g) The addition of iron to water also produces chemical precipitates which remove pollutants by the same two primary mechanisms previously discussed for aluminum. Removal of suspended solids, algae, phosphorus, heavy metals, and bacteria occurs primarily by enmeshment and adsorption onto iron hydroxide precipitate according to the following net reaction: Fe H 2 O Fe(OH) 3(s) + 3H 3 O + Removal of additional dissolved phosphorus occurs as a result of formation of FePO 4 by: 19

99 Final Report Chemical Treatment Fe +3 + HnPO 4 n-3 FePO 4(s) + nh + Immediately after addition to water, Al +3 and Fe +3 cations undergo hydration reactions in aqueous systems which are governed by a variety of factors such as the presence of other inorganic ligands, concentration of the metal ion, and ph of the solution. The hydrolytic reactions are so rapid that raw metal ions do not exist, and ionic species occur as a variety of soluble monomeric, dimeric, and polymeric hydroxo-metal complexes. A listing of significant hydrolytic reactions and equilibrium constants for aluminum, iron, and calcium reactions are given in Table 2-3. Both Al and Fe are ampoteric and capable of forming both cationic and anionic complexes. REACTION NUMBER TABLE 2-3 HYDROLYTIC REACTIONS AND CONSTANTS FOR ALUMINUM, IRON, AND CALCIUM 25 o C REACTION LOG K eq 1 Al +3 + H 2 O = AlOH +2 + H Al H 2 O = Al(OH) H Al H 2 O = Al(OH) 3(aq) + 3H Al H 2 O = Al(OH) H Al H 2 O = Al 2 (OH) H Al H 2 O = Al 3 (OH) H Al H 2 O = Al 13 O 4 (OH) H Al(OH) 3(s) + 3H + = Al H 2 O Al(OH) 3 + 3H + = Al H 2 O amorph Al OH - = Al(OH) 3(s) Fe +3 + H 2 O = FeOH +2 + H Fe H 2 O = Fe(OH) H Fe H 2 O = Fe(OH) 3(aq) + 3H + < Fe H 2 O = Fe(OH) H Fe H 2 O = Fe 2 (OH) H Fe(OH) (s) + 3H + = Fe H 2 O (am) Fe(OH) (s) + 3H + = Fe H 2 O Fe OH - = Fe(OH) 3(s) CaHPO 4(s) = Ca +2 + HPO Ca 4 H(PO 4 ) 3(s) = 4Ca PO H Ca 10 (PO 4 ) 6 (OH) 2(s) = 10Ca PO OH Ca 10 (PO 4 ) 6 (F) 2(s) = 10Ca PO F Ca 10 (PO 4 ) 6 (OH) 2(s) + 6H 2 O = 4[Ca 2 (HPO 4 )(OH) 2 ] + 2Ca HPO

100 Log [Conc.] Log [Conc.] Final Report Chemical Treatment A solubility diagram for freshly precipitated and aged Al(OH) 3 floc is given in Figure 2-1. The equilibrium solubility of aluminum is primarily a function of ph and age of the floc. Freshly precipitated Al(OH) 3 floc has a minimum solubility of approximately 10-5 M which occurs in the ph range of However, over a period of several months, the alum floc ages, eventually forming gibbsite, with a minimum solubility of approximately 10-9 M. As this aging process occurs, the ph range of minimum solubility shifts slightly into the range of approximately Freshly Precipitated Al(OH) 3 Aged Al(OH) Al(OH) 3(s) Al(OH) 3(s) -6-6 Al 2 (OH) 2 +4 Al 13 (OH) Al 2 (OH) Al Al +3 Al(OH) ph ph Figure 2-1. Solubility Diagram for Freshly Precipitated and Aged Al(OH) 3. (Adapted from Snoeyink and Jenkins, 1980). A solubility diagram for freshly precipitated Fe(OH) 3 floc is given on Figure 2-2. The minimum solubility for this floc is approximately 10-9 M which occurs in the ph range of approximately The stability of the floc decreases substantially and the solubility of Fe +3 increases substantially at ph values both lower and higher than this range. Unlike Al(OH) 3, Fe(OH) 3 does not undergo a significant aging process or shift in solubility characteristics over time. A solubility diagram for calcium phosphate compounds is given on Figure 2-3. The minimum solubility for calcium phosphate compounds is approximately M which occurs in the ph range of approximately The stability of the floc decreases substantially and the solubility of Ca +2 increases substantially at ph values less than 8. Unlike AlOH 3, calcium phosphate 21

101 Log [Fe] Final Report Chemical Treatment compounds do not undergo a significant aging Figure process 2-2or shift in solubility characteristics over time Fe(OH) 3(s) ph Figure 2-2. Solubility Diagram for Amorphous Fe(OH) 3. (Adapted from Faust and Aly, 1998). 22

102 Log [Soluble P] Final Report Chemical Treatment Figure CaHPO Ca 4 H(PO 4 ) 3-8 Ca 10 H(PO 4 ) 6 (OH) 2-10 Ca 10 H(PO 4 ) 6 (F) ph Figure 2-3. Solubility Diagram for Calcium Phosphate Complexes. (Adapted from Stumm and Morgan, 1981). In addition to phosphorus removal by absorption onto metal hydroxides, iron and aluminum compounds can also precipitate dissolved orthophosphorus directly as metal phosphate compounds. Solubility diagrams for ferric and aluminum phosphate are given on Figure 2-4. The minimum solubility for ferric phosphate appears to be approximately M which occurs at a ph of 4-5. The minimum solubility for aluminum phosphate (AlPO 4 ) (s) is approximately M which occurs at a ph value of approximately These diagrams appear to suggest that alum will provide a lower equilibrium concentration when used for coagulating waters with high concentrations of orthophosphate. 23

103 Log [Fe] Log [Al] Final Report Chemical Treatment -3 Figure FePO 4(s) -4 AlPO 4(s) Fe(OH) 3(s) Al(OH) 3(s) ph ph Figure 2-4. Solubility Diagrams for Ferric and Aluminum Phosphate. Physical Factors Affecting Coagulation Many factors are capable of affecting the coagulation process using metal salts. The most significant factors include: (a) coagulant dosage, (b) ph, (c) natural color concentration, (d) competing ions in solution, (e) mixing effects, and (f) temperature. In general, the performance efficiency of metal salt coagulants increases in a non-linear fashion with increases in coagulant dose, provided that a relatively neutral ph is maintained during the process. Although the performance efficiency increases with increasing coagulant dose, the additional removal efficiency achieved begins to level off and become asymptotic at elevated doses. The impact of ph on the coagulation process has been discussed in previous sections. In general, the coagulation process is maximized, and residual metal concentrations minimized, when the coagulated water is maintained within the ph range of minimum solubility for the applied coagulant. For alum, this ph zone is approximately 6-7, while for iron the minimum solubility occurs in the ph range of 8-10, and in the ph range of for calcium. A considerable amount of information has been developed concerning the chemical nature of 24

104 Final Report Chemical Treatment organic color in natural waters. Organic color generally has the physical property of a negatively charged colloid with particle sizes ranging from m. When the dissolved organic carbon (DOC) in a water is low, the formation of humic-aluminum precipitates is favored which often have poor settling characteristics. However, when the DOC is high, the precipitation process favors formation of Al(OH) 3 which is a more rapidly settling precipitate. Competing ions in solution can substantially impact the kinetics of the coagulation process. Anions such as sulfate have long been known to suppress the charge reversal process which is primarily responsible for formation of settleable floc material. Also, the presence of divalent ions such as Ca +2 and Mg +2 have been shown to enhance the coagulation process. Temperature may also have a significant impact on the coagulation process. Under cold temperatures, floc formation and the removal efficiency achieved using metal salts for coagulation decreases substantially. Colder temperatures often require a change in coagulant or change in dose to maintain acceptable settling characteristics and removal efficiencies. However, under conditions commonly observed within the State of Florida, temperature is generally an insignificant parameter impacting coagulation processes. Impacts of Redox Potential Aluminum and calcium do not exhibit alternative oxidation states in the natural environment, and both aluminum and calcium compounds are immune to changes in redox potential within the collected floc. Compounds absorbed onto aluminum or calcium floc are equally stable under aerobic or anoxic conditions. However, iron compounds can exhibit several different electron configurations, the most common of which involve the ferric (Fe +3 ) and ferrous (Fe +2 ) ions. Under oxidized conditions, indicated by redox potentials in excess of 200 mv (Eh), iron compounds are predominantly present in the ferric ion state. Compounds formed with ferric ions are highly insoluble under aerobic conditions. When the redox potential drops below 200 mv and reduced conditions dominate, the ferric ion accepts an electron and is converted into a highly soluble ferrous ion, as shown in the following reaction: Fe +3 + e - Fe +2 Any contaminants which have been adsorbed onto the iron floc will be released as the floc dissolves under the reduced conditions. Therefore, iron compounds should only be used for coagulation in processes where aerobic conditions can be assured at all times. Iron compounds may not be suitable for use in systems where the floc is collected and stored in a pond for long periods of time in submerged conditions. A summary of iron solubility as a function of ph and redox potential is given in Table 2-4. For example, at a ph of 7.0, iron is 10,000,000,000,000 times more soluble under reduced conditions than under highly oxidized conditions. 25

105 Final Report Chemical Treatment TABLE 2-4 SOLUBILITY OF IRON SPECIES AS A FUNCTION OF ph AND REDOX POTENTIAL SPECIES ph Fe Fe +2 E h = 800 mv E h = 600 mv E h = 400 mv E h = 300 mv E h = 250 mv E h = 200 mv E h = 0 mv 300,000 10, ph Impacts One of the most significant issues involved in the selection and use of chemical coagulants is the potential for either consumption or addition of alkalinity and the resulting impacts on ph. A comparison of alkalinity addition or consumption during coagulation with common treatment chemicals is given in Table 2-5. When iron or aluminum coagulants are used, alkalinity is consumed as a result of the coagulation process which can result in a decrease in solution ph, depending upon the applied coagulant dose and the available buffering capacity of the source water. As seen in Table 2-5, the alkalinity consumption during coagulation is higher with ferric coagulants than with aluminum sulfate, aluminum chloride, or PACl. This suggests that at equal doses the addition of ferric chloride will have a more significant impact on ph than would be observed using aluminum-based coagulants. In contrast, alkalinity is added to the source water during coagulation with alkaline coagulants, such as lime, sodium hydroxide, or sodium aluminate. 26

106 Final Report Chemical Treatment TABLE 2-5 ALKALINITY ADDITION OR CONSUMPTION DURING COAGULATION WITH COMMON TREATMENT CHEMICALS (Lind, 1997) CHEMICAL (BASIS) CHANGE IN ALKALINITY (ppm as CaCO 3 per ppm Product) Ferric Chloride/Sulfate (liquid) Aluminum Sulfate (dry basis) Aluminum Chloride (liquid) PACl (liquid) -0.3 to (varies with product) Lime (dry) Sodium Hydroxide (dry) Soda Ash (sodium carbonate) (dry) Sodium Bicarbonate (dry) Sodium Aluminate (liquid) +0.4 to +0.6 (varies with product) APPLICATIONS AND SUCCESS STORIES At Least Fifty Five Facilities in the State Environmental Research & Design, Inc. (ERD) pioneered the concept of using chemical coagulants for treatment of stormwater and tributary inflows during the mid-1980s. The first system designed for chemical treatment of stormwater was constructed on Lake Ella, Tallahassee, during This system injects liquid alum into the incoming stormwater on a flow-proportioned basis. The alum forms inert precipitates of Al(OH) 3 and AlPO 4, which sorb phosphorus, suspended solids, heavy metals, organic compounds, and bacteria as it settles from the water column into the lake sediments. This system provided a cost effective and highly efficient method for treatment of stormwater runoff in an urban setting. Since that time, ERD has designed and constructed more than 55 additional alum treatment systems in urban settings and conducted literally hundreds of laboratory jar tests to evaluate treatment feasibility for a wide range of water characteristics collected throughout the State of Florida. ERD has also conducted FDEP sponsored research to address a variety of potential issues related to chemical treatment, such as removal efficiencies, reaction kinetics, floc generation rates, floc characteristics, floc disposal, benthic and ecological impacts, floc stability, and treatment costs. During 1988, ERD conducted an evaluation of the feasibility of using chemical treatment to reduce phosphorus loadings from agricultural discharges into Lake Apopka. A pilot system was constructed which confirmed the ability of the process to remove phosphorus from pumped agricultural discharges. During 1995, ERD evaluated, designed, and constructed an alum treatment system for the primary agricultural inflow to Lake Apopka (30,000 gpm), which is still in operation today. Since that time, ERD has designed, and in some cases, constructed, 5 additional large scale treatment systems for tributary and agricultural discharges, one of which is located in the Northern Okeechobee Basin. The most recent system, located in the Lake Apopka 27

107 Final Report Chemical Treatment basin, is capable of continuously treating up to 300 cfs. Initial Testing and Evaluation Once alum has been identified as an option in a stormwater management or retrofit project, extensive laboratory testing must be performed to verify the feasibility of alum treatment and to establish process design parameters. The feasibility of alum treatment for a particular stormwater stream is typically evaluated in a series of laboratory jar tests conducted on representative runoff samples collected from the project watershed area. This laboratory testing is an essential part of the evaluation process necessary to determine design, maintenance, and operational parameters such as the optimum coagulant dose required to achieve the desired water quality goals, chemical pumping rates and pump sizes, the need for additional chemicals to buffer receiving water ph, post-treatment water quality characteristics, floc formation and settling characteristics, floc accumulation, annual chemical costs and storage requirements, ecological effects, and maintenance procedures. In addition to determining the optimum coagulant dose, jar tests can also be used to determine floc strength and stability, required mixing intensity and duration, and determine design criteria for settling basins. Since 1986, Environmental Research & Design, Inc. (ERD) has performed literally hundreds of laboratory flocculation jar tests to evaluate the effectiveness of alum for reducing concentrations of common constituents in stormwater runoff collected from a wide range of urban land use activities. A summary of mean removal efficiencies achieved during alum treatment of stormwater runoff for typical stormwater pollutants is given in Table 2-6. Removal efficiencies are summarized for alum treatment of stormwater runoff at doses of 5, 7.5 and 10 mg Al/liter, as well as stormwater samples which were allowed to settle under quiescent conditions for a period of 24 hours to simulate removal efficiencies which would be achieved using a wet or dry detention stormwater treatment basin for comparison purposes (Harper, et al., 1998). TABLE 2-6 TYPICAL REMOVAL EFFICIENCIES FOR ALUM TREATED STORMWATER RUNOFF PARAMETER SETTLED ALUM DOSE (mg/l as Al) WITHOUT ALUM 5 mg/l 7.5 mg/l 10 mg/l Diss. Organic N Particulate Nitrogen Total Nitrogen Diss. Orthophosphorus Particulate P Total P Turbidity TSS BOD Total Coliform Fecal Coliform

108 Final Report Chemical Treatment As seen in Table 2-6, alum treatment of stormwater runoff consistently achieves an 85-95% reduction in total phosphorus, 20-70% reduction in total nitrogen, 95-99% reduction in turbidity and TSS, and 96-99% reduction in fecal coliform bacteria. Removal efficiencies of 50-90% are also achieved for heavy metals. The minimum tested dose of 5 mg Al/liter is generally considered to be the minimum dose necessary to achieve acceptable floc settling characteristics. Removal efficiencies for measured constituents appear to increase slightly with increasing alum dose. In general, removal efficiencies achieved using alum are substantially greater than those achieved using settling alone. Removal of total phosphorus in alum treated stormwater occurs by direct precipitation of orthophosphorus as aluminum phosphate (AlPO 4 ), as well as enmeshment of particulate phosphorus by incorporation into Al(OH) 3 floc. Removal of nitrogen species occurs primarily as a result of precipitation of particulate nitrogen and dissolved organic nitrogen, since alum treatment generally does not affect measured concentrations of ammonia or nitrate. As seen in Table 2-6, alum treatment removal efficiencies for nitrogen can be highly variable. In general, alum treatment has only a minimal effect on concentrations of ammonia and virtually no impact on concentrations of NO x in stormwater runoff. Removal of dissolved organic nitrogen species can also be highly variable, depending upon molecular size and structure of the organic compounds. The only nitrogen species which can be removed predictably is particulate nitrogen. As a result, removal efficiencies for total nitrogen are highly dependent upon the nitrogen species present, with higher removal efficiencies associated with runoff containing large amounts of particulate and organic nitrogen and lower removal efficiencies for runoff flows which contain primarily inorganic nitrogen species. Selection of the "optimum" dose often involves an economic evaluation of treatment costs vs. desired removal efficiencies. System Configurations In a typical alum stormwater treatment system, alum is injected into the stormwater or tributary flow on a flow-proportioned basis so that the same dose of alum is added regardless of the discharge rate. A variable speed chemical metering pump is typically used as the injection pump. If the initial laboratory testing indicates that the addition of alum to the target runoff flow will reduce ph levels to undesirable levels, a buffering agent, such as sodium aluminate (Na 2 Al 2 O 4 ) or sodium hydroxide (NaOH) can be injected along with the alum to maintain desired ph levels. A separate metering system and storage tank will be necessary for the buffering agent. The operation of each injection pump is regulated by a flow meter device attached to each incoming stormwater line to be treated. Measured flow from each stormwater flow meter is transformed into a 4-20 ma electronic signal which instructs the metering pump to inject alum according to the measured flow of runoff discharging through each individual stormsewer line. Mixing of the alum and stormwater occurs as a result of turbulence in the stormsewer line. If sufficient turbulence is not available within the stormsewer line, artificial turbulence can be generated using aeration or physical stormsewer modifications. Since alum addition is regulated by the rate of flow in the stormsewer line, the treatment system is capable of treating stormwater 29

109 Final Report Chemical Treatment as well as dry weather baseflow. A series of rate experiments were conducted by Harper (1990) to evaluate the time required for dissolution of alum floc. Since Al +3 can be a potentially toxic species, floc formation should be complete prior to discharging the treated stormwater into the receiving waterbody. It was determined that floc formation is complete, although on a microscopic scale, and Al +3 is virtually removed from the water column, in seconds after alum addition. Therefore, alum injection locations are carefully selected to allow a minimum of seconds of travel time in the stormsewer line after alum addition prior to reaching the receiving waterbody. Mechanical components for an alum stormwater treatment system include chemical metering pumps and stormsewer flow meters and electronic controls which are typically housed in a central facility constructed as an above-ground or below-ground structure. A fiberglass storage tank is typically used for bulk alum storage. Alum feed lines and electrical conduits are run from the central facility to each point of alum addition and flow measurement. Alum injection points can be located as far as 3000 ft or more from the central pumping facility. The capital costs of constructing an alum stormwater treatment system do not increase substantially with increasing size of the drainage basin which is treated. As a result, alum treatment has become increasingly popular in large regional treatment systems. Prior to 1998, many of the constructed alum stormwater treatment systems allowed the generated aluminum floc to settle directly in the receiving waterbody. An example of this type of system is the Lake Howard alum stormwater treatment system located in downtown Winter Haven. This system provides alum injection to seven separate stormsewer systems which drain approximately 261 acres of commercial and single-family land use adjacent to Lake Howard. An overview of the Lake Howard alum stormwater treatment system is given on Figure 2-5. A single chemical metering pump is used to inject alum to each of the seven points of injection, with flow control valves used to regulate the amount of alum added at each injection point. Mixing of the alum and stormwater occurs within the stormsewer system, and the generated alum floc discharges directly into Lake Howard. The electrical components and pumping equipment are contained within a small equipment building constructed on vacant land adjacent to the Lake. One of the earliest stormwater treatment systems was constructed on Lake Lucerne, a 29-acre urban lake located near downtown Orlando. Lake Lucerne receives untreated stormwater runoff from a 267-acre watershed which includes much of downtown Orlando. An alum stormwater treatment system was designed which injects liquid alum into six primary stormsewer systems discharging to Lake Lucerne which contribute approximately 90% of the annual runoff inputs to the lake. Mechanical components for the Lake Lucerne alum treatment system are housed in an underground vault beneath an elevated expressway and required no land purchase for construction. The floc generated during the coagulation process discharges directly into the lake. Photographs of Lake Lucerne and the underground pump and control building are given on Figure

110 Final Report Chemical Treatment Equipment Building Equipment Building Underground Alum Storage Tank Underground Alum Storage Tan Floc Discharge to Lake Alum Inj Equipment Alum Injection Equipment Figure 2-5. Overview of the Lake Howard Alum Stormwater Treatment System. Virtually all of the alum stormwater treatment systems permitted after 1998 provide mechanisms for collection and removal of the generated floc. One of the first systems designed for automatic collection and removal of the alum floc is referred to as the Gore Street treatment system which provides alum treatment for a 250-acre watershed in downtown Orlando which discharges into Clear Lake. An overview of the Gore Street alum stormwater treatment and floc collection system is given on Figure 2-7. The mechanical components for the system are housed in a small prefabricated concrete building located adjacent to the point of injection. Alum is injected into a large 8-ft x 10-ft CBC, and the generated floc is collected in an expanded portion of the channel 31

111 Final Report Chemical Treatment which connects the box culvert to Clear Lake. A semi-permeable fabric is used to collect the alum floc while allowing water to pass through Figure the 2-6 fabric. The generated floc is removed periodically by pumping directly into the adjacent sanitary sewer system. Figure 2-6 Pump control panels Chemical metering pumps Flow meter control panels Figure 2-6. Photographs of Lake Lucerne and the Underground Pump and Control Building. 32

112 Final Report Chemical Treatment Equipment Building Equipment Building Floc Disposal System Floc Disposal System Floc Pumping Equipment In-Line Floc Trap In-line Floc Trap In-Line Floc Trap In-line Floc Trap Permeable Fabric Figure 2-7. Overview of the Gore Street Alum Stormwater Treatment System. One of the larger alum stormwater treatment systems was constructed in the City of Largo to provide treatment for a 1158-acre watershed which discharges through a canal into Tampa Bay. An overview of the Largo regional alum treatment system is given on Figure 2-8. A drivable diversion weir was constructed across the channel to divert the canal water into an underground box culvert. The flow rate through the box culvert is measured using a flow meter, and alum is injected according to the rate of stormwater flow. Mechanical and electrical components for the 33

113 Final Report Chemical Treatment alum injection system are housed in an adjacent concrete block building. The alum floc is discharged into a floc settling pond, and the clear water discharges through an outfall structure back into the original channel. Collected floc is removed periodically from the settling pond using a series of underwater sumps which are connected to the adjacent sanitary sewer lift station. Figure 2-8 Drivable Drainage Diversion Weir Alum Injection Building Canal Flow Diverted Into Box Culvert Flow Figure 2-8 Elevated Wooden Boardwalk Wetland Enhancement Floating Dock Floc Settling Pond Inflow Outflow 15 Acre Hardwood Wetland Enhancement Paved Walking Path Figure 2-8. Overview of the Largo Regional Alum Treatment Facility. 34

114 Final Report Chemical Treatment The largest alum stormwater treatment system to date is located along the Apopka-Beauclair Canal which extends between Lake Apopka and Lake Beauclair in Central Florida. This canal carries discharges from Lake Apopka, a 30,000-acre shallow hypereutrophic lake, into Lake Beauclair which forms the headwaters of the Harris Chain-of-Lakes. Inflow from the Apopka-Beauclair Canal into Lake Beauclair is thought to be the single largest source of phosphorus loadings to the Harris Chain-of-Lakes. The Apopka-Beauclair Canal Nutrient Reduction Facility (NuRF) is designed to provide alum treatment for the canal discharges prior to reaching Lake Beauclair. A schematic of the NuRF Facility is given on Figure 2-9. Discharge rates and water level elevations in the Apopka-Beauclair Canal are regulated by the Apopka-Beauclair Canal lock and dam. The NuRF Facility uses the difference in water level elevations between upstream and downstream portions of the canal to force the upstream canal water into two parallel treatment basins. Liquid alum is added upstream of the point of inflow into the treatment basins, and the generated floc settles onto the bottom of the ponds. These basins are designed to allow treatment of up to 300 cfs while still providing a minimum detention time of three hours for capture of the floc material. Treated discharges from the ponds enter a small canal which conveys the treated water downstream of the lock and dam structure where it ultimately reaches Lake Beauclair. Flow in excess of 300 cfs, which rarely occurs, will be allowed to bypass the treatment system. To Lake Beauclair Lock and Dam Structure Figure 2-9 From Lake Apopka Outflow Canal Inflow Canal 300 cfs max. Treatment Pond 1 200,000 gal Floc mixing tank Treatment Pond 2 Dried floc storage area Alum Pumping & Control Bldg. 62,000 gal alum storage tanks Floc dewatering facility Figure 2-9. Overview of the Lake County NuRF Facility. 35

115 Final Report Chemical Treatment Approximately 1-2 times each year, depending upon treated flow rates, floc removal will be necessary from the two settling ponds. This removal will be achieved using an automated dredging system constructed as part of these ponds. This system will automatically dredge the accumulated floc from the bottom of the pond and pump the dredge slurry to a large centrifuge located in the adjacent floc processing building. The centrifuge will decrease the water content of the sludge to approximately 40%, so that it can be hauled to the adjacent floc drying area. The floc drying area consists of an elevated area constructed on permeable soils where the floc will continue to dry naturally. It is anticipated that the dry floc will be used either as landfill cover or by the St. Johns River Water Management District as a soil amendment for various Lake Apopka restoration projects. The alum floc still contains considerable uptake capacity for phosphorus and other species and can be used to reduce phosphorus release from flooded farm lands which are converted to water quality treatment areas. The NuRF Facility contains storage capabilities for approximately 124,000 gallons of alum to meet chemical demand under high flow conditions. At the maximum design treatment rate of 300 cfs, the facility will utilize approximately eight tanker loads (4500 gallons) of alum each day. The construction cost for the facility was approximately $7.5 million, with an anticipated annual alum consumption in excess of 1 million gallons. A recent innovation in alum treatment systems is currently being constructed at Lake Seminole in Pinellas County. Lake Seminole is a large eutrophic urban lake which has been hydraulically impacted by construction of the adjacent Seminole Bypass Canal. To increase flushing within the lake, water from the Seminole Bypass Canal is pumped into a linear alum treatment system at a constant flow rate of 10 cfs. The generated floc settles onto the bottom of a trough-type collection system, and the treated water is then discharged into lake Seminole to increase flushing and provide a source of clean water. The collected floc settles onto the bottom of the trough and is removed automatically on a daily basis using a series of control valves and floc collection pumps. The collected floc is then discharged to the sanitary sewer system for disposal. A schematic of the Lake Seminole Bypass Canal treatment system is given on Figure Alum treatment has also been evaluated for use in reducing nutrient concentrations in agricultural runoff. Harper (1987) performed an extensive study to evaluate the effectiveness of alum for reducing nutrient concentrations in agricultural runoff from the Central Florida area. The evaluated farm areas were utilized primarily for row crops which were grown in high organic muck and peat type soils. Runoff generated from these areas was found to contain high levels of color, with large portions of inorganic and organic nutrient forms. The dominant nitrogen species was found to be dissolved organic nitrogen, while the dominant phosphorus species was dissolved orthophosphorus. Typical changes in water quality characteristics resulting from alum treatment of agricultural runoff are summarized in Table 2-7. In general, alum treatment resulted in slight reductions in ph and alkalinity in the treated water, with corresponding increases in specific conductivity. Inorganic nitrogen species were relatively unaffected by the treatment process, with the majority of total nitrogen removal occurring as a result of reduction in concentrations of organic nitrogen. Alum treatment was observed to be extremely effective in reducing concentrations of dissolved orthophosphorus with more than 90% removal achieved at alum doses in excess of 10 mg Al/liter. Alum treatment was also effective in reducing concentrations of TSS and BOD, with approximately 50% removal for these parameters at alum doses in excess of 10 mg Al/liter. In general, efficient 36

116 Final Report Chemical Treatment Figure 2-10 removal of phosphorus Lake species Seminole from agricultural Bypass runoff Canal using Treatment alum required System doses which were approximately two times greater than the doses necessary to substantially reduce nutrient concentrations Seminole Lake urban Seminole runoff. Bypass Canal Bypass Canal Treatment System Figure 2-10 Alum Seminole pumping Bypass Canal and control bldg. Inflow Treatment System 10 cfs Alum pumping and control bldg. Inflow Treatment System 10 cfs Lake Seminole Treated Discharge Treated Discharge Lake Seminole First system which is totally First system automated which is totally automated 25 ft. 25 ft. to sanitary sewer Floc collection Cross-section of Treatment System system discharge to sanitary sewer Figure Cross-section Schematic of the of Lake Treatment Seminole System Bypass Canal Treatment System. TABLE 2-7 TYPICAL CHANGES IN WATER QUALITY CHARACTERISTICS RESULTING FROM ALUM TREATMENT OF AGRICULTURAL RUNOFF 1 PARAMETER UNITS RAW WATER 37 Floc collection system discharge ALUM TREATED (Dose in mg Al/liter) ph s.u Alkalinity mg/l Specific Conductivity μmho/cm NH 3 -N μg/l NO x -N μg/l Organic Nitrogen μg/l Total Nitrogen μg/l Orthophosphorus μg/l Total Phosphorus μg/l TSS mg/l BOD mg/l Harper (1987)

117 Final Report Chemical Treatment During 1997, ERD designed, constructed, and operated a large-scale alum injection system to treat pumped discharges into Lake Apopka which originated from muck farming areas adjacent to the lake. This system was constructed as a pilot-scale operation to demonstrate the efficacy of alum injection for reducing dissolved phosphorus concentrations. The treatment system was constructed in a 50-ft wide channel upstream of the point of intake for a 30,000 gpm electric pump. This pump is utilized on a routine basis to pump accumulated water from the farming area into Lake Apopka to avoid flooding within the farming area. The system was designed to reduce orthophosphorus concentrations by approximately 80% from initial orthophosphorus concentrations in the range of ppb. The treatment system was extremely effective in reducing concentrations of dissolved orthophosphorus, with an alum dose of approximately mg/l required for this removal. However, due to the elevated concentrations of dissolved orthophosphorus and the resulting formation of significant quantities of AlPO 4 precipitate, the generated floc exhibited relatively poor settling characteristics with a minimum detention period of approximately hours required for clarification. This pilot system was later modified to a permanent system which is still in operation today. The floc precipitate generated during this process was originally allowed to discharge into Lake Apopka although a settling pond is used for collection under current conditions. Water Quality Improvements In general, construction and operation of alum stormwater treatment systems have resulted in significant improvements in water quality for treated waterbodies. The degree of observed improvement in water quality is directly related to the percentage of annual hydraulic inputs treated by the alum stormwater treatment system. A comparison of pre- and post-modification water quality characteristics for typical alum stormwater treatment systems, including Lake Ella and Lake Dot (which provide treatment for approximately 95-96% of the annual hydraulic inputs entering these lake systems), and Lake Osceola (which provides treatment for only 9% of the annual hydraulic inputs entering this lake system) is given in Table 2-8. Lake Dot is located in Orlando, Florida, and Lake Osceola is located in Winter Park, Florida. In general, operation of the alum stormwater treatment systems have resulted in a decline in ph within each of the three waterbodies, with a reduction of approximately 1 unit in Lake Ella and 0.6 units in Lake Osceola. A ph reduction of only 0.1 unit was observed for the Lake Dot treatment system which injects both alum and sodium aluminate, an alkaline form of alum, to control ph levels within the lake. Significant improvements in dissolved oxygen were also observed in both Lake Ella and Lake Dot. Alum treatment of stormwater runoff resulted in a 78% reduction in total nitrogen concentrations in Lake Ella, with a 55% reduction in Lake Dot and a 4% reduction in Lake Osceola where only a small portion of the annual hydraulic inputs were treated. The majority of the total nitrogen removal occurred as a result of reducing concentrations of dissolved organic nitrogen and particulate nitrogen since alum is generally ineffective in reducing concentrations of inorganic nitrogen species, such as ammonia or nitrate. Alum stormwater treatment resulted in a substantial reduction in measured concentrations of orthophosphorus and total phosphorus in each of the three lake systems, with total removals of 38

118 Final Report Chemical Treatment 89%, 93% and 30% for Lake Ella, Lake Dot and Lake Osceola, respectively. Alum stormwater treatment also reduced in-lake concentrations of BOD in each of the three lake systems, with a reduction of 93% in Lake Ella and 84% in Lake Dot. TABLE 2-8 COMPARISON OF PRE- AND POST-MODIFICATION WATER QUALITY CHARACTERISTICS FOR TYPICAL ALUM STORMWATER TREATMENT SYSTEMS PARAMETER UNIT S LAKE ELLA (Tallahassee, FL) LAKE DOT (Orlando, FL) LAKE OSCEOLA (Winter Park, FL) Before ( ) After (1/88-5/90) Before ( ) After (3/89-8/91) Before (6/91-6/92) After (2/93-12/96) # of Samples ph s.u Diss. O 2 (1 minute) mg/l Total N g/l Total P g/l BOD mg/l Chlorophyll-a mg/m Secchi Disk Depth m 0.5 > 2.2 < Diss. Al g/l Florida TSI Value (Hypereutrophic) 47 (Oligotrophic) 86 (Hypereutrophic) 42 (Oligotrophic) 61 (Eutrophic) 56 (Mesotrophic) Lake Area ac 5.9 ac 55.4 ac Watershed Area ac 305 ac 153 ac Percent of Annual Hydraulic Inputs Treated % Alum stormwater treatment results indicate that alum may be extremely effective in reducing primary productivity in receiving waterbodies, as indicated by concentrations of chlorophyll-a, with a reduction of 97% in Lake Ella, 89% in Lake Dot and 13% in Lake Osceola. Reductions in measured concentrations of chlorophyll-a occur as a result of enmeshment and precipitation of algal particles within the water column of the lake by alum floc as well as phosphorus limitation created by low levels of available phosphorus in the water column. Substantial increases in Secchi disk depth were observed in Lake Ella and Lake Dot, and to a lesser extent in Lake Osceola, with improvements of 340% in Lake Ella, 212% in Lake Dot and 9% in Lake Osceola. Based upon the Florida TSI Index (Brezonik, 1984), Lake Ella and Lake Dot have been converted from hypereutrophic to oligotrophic status, with a conversion from eutrophic to mesotrophic in Lake Osceola. A graphical history of total phosphorus concentrations in Lake Lucerne, which was retrofitted with an alum stormwater treatment system in June 1993 that provides treatment for approximately 82% of the annual runoff inputs into the lake, is given in Figure Prior to 39

119 Total Phosphorus (µg/l) Testing / Startup System Offline Final Report Chemical Treatment construction of the alum stormwater treatment system, total phosphorus concentrations in Lake Lucerne fluctuated widely, with a mean concentration of approximately 100 g/l. Following start-up of the alum treatment system, total phosphorus concentrations began to decline steadily, reaching equilibrium concentrations of approximately g/l. A slight increase in total phosphorus concentrations is observed during the last half of 1995 when the system was off-line due to lightning damage. When system operation Figure resumed 2-11 in June 1996, total phosphorus concentrations returned to equilibrium values of approximately 20 g/l. Mean residence time in Lake Lucerne is approximately 105 days. Total Phosphorus Before Alum During System Operation During System Operation Date Figure Trends in Total Phosphorus Concentrations in Lake Lucerne, Before and After Alum Treatment of Stormwater Runoff. In general, measured concentrations of heavy metals have been extremely low in value in all waterbodies retrofitted with alum stormwater treatment systems, with no violations of heavy metal standards observed in any of these lake systems. Measured levels of dissolved aluminum have also remained low in each lake system. Mean dissolved aluminum concentrations for Lake Ella, Lake Dot and Lake Osceola have averaged 44 g/l, 65 g/l and 51 g/l, respectively. Although there is no standard for dissolved aluminum in the State of Florida, the U.S. EPA has recommended a long-term average of 87 g/l for protection of all species present in the U.S. The solubility of dissolved aluminum is regulated almost exclusively by ph. As long as the ph of the treated water can be maintained in the range of during the treatment process, dissolved aluminum concentrations will remain at minimal levels. During 2002, construction was completed on a joint cooperative project between the City of Largo and the Southwest Florida Water Management District (SWFWMD) to construct a 40

120 Final Report Chemical Treatment regional alum stormwater treatment facility to treat pollutant loads discharging from a 1158-acre watershed which discharges directly to Boca Ciega Bay and Tampa Bay. Photographs of the Largo regional alum treatment facility are given on Figure 2-8. Performance efficiency monitoring of the treatment facility was conducted by ERD from September 2002-February Monitoring was conducted of the raw runoff inflow prior to alum addition and at the discharge from the alum floc settling pond. A summary of the changes in phosphorus concentrations observed during field monitoring program is given on Figure The alum treatment facility resulted in substantial reductions in measured concentrations of dissolved orthophosphorus, dissolved organic phosphorus, particulate phosphorus, and total phosphorus. During the monitoring program, the treatment system achieved a removal efficiency of approximately 85% for total phosphorus, 88% for TSS, and 37% for total nitrogen. The observed removal efficiencies achieved by the system were substantially in excess of the predicted annual load reductions predicted during the preliminary evaluation phase. This system has resulted in a significant reduction in nutrient loadings discharging to Tampa Bay. According to SWFWMD, the load reductions achieved by the Largo project represent approximately 60% of the overall load reduction goals to Tampa Bay. 250 Diss. Ortho-P Diss. Organic P Particulate P Total P Phosphorus Concentration (µg/l) Inflow Outflow 0 Figure Fate of Phosphorus Species in the Largo Regional Stormwater Facility. 41

121 Final Report Chemical Treatment Floc Accumulation Laboratory investigations have been conducted on stormwater runoff collected from a wide range of land uses typical of urban areas to quantify the amount of alum floc generated as a result of alum treatment of stormwater runoff at various treatment doses. After initial formation, alum floc appears to consolidate rapidly for a period of approximately 6-8 days, compressing approximately 20% of the initial floc volume. Additional consolidation appears to occur over a settling period of approximately 30 days, after which collected sludge volumes appear to approach maximum consolidation (Harper, 1990). Estimates of maximum anticipated sludge production, based upon literally hundreds of laboratory tests involving coagulation of stormwater runoff with alum at various doses, and a consolidation period of approximately 30 days, is given in Table 2-9. At alum doses typically used for treatment of stormwater runoff, ranging from 5-10 mg Al/liter, sludge production is equivalent to approximately % of the treated runoff flow. Sludge production values listed in Table 2-9 reflect the combined volume generated by alum floc as well as solids originating from the stormwater sample. TABLE 2-9 ANTICIPATED PRODUCTION OF ALUM SLUDGE FROM ALUM TREATMENT OF STORMWATER AT VAROUS DOSES AFTER A 30-DAY CONSOLIDATION PERIOD ALUM DOSE SLUDGE PRODUCTION (mg/l as Al) As Percent of Per 1000 Per 10 6 Treated Flow m 3 Treated Gallons Treated m ft m ft m ft 3 After collection, alum floc undergoes a drying process similar to a wastewater sludge. Photographs of a typical drying process for alum floc are given on Figure When fresh floc is collected, it generally dewaters rapidly, with a cracked cake forming in approximately one week. Complete dewatering of the sludge generally requires approximately days, depending upon weather conditions. A volume reduction of approximately 95% is achieved in the dried sludge compared with the fresh floc. 42

122 Final Report Chemical Treatment Figure 2-12 Fresh Floc After ~ 1 week After ~ 30 days Figure Typical Drying Process for Alum Floc. Floc Characteristics and Stability A substantial amount of data has been collected by ERD regarding the physical and chemical characteristics of alum residual generated as a result of alum treatment of tributary inflows and stormwater runoff. One of the most recent studies was conducted as part of the preliminary evaluation phase for the Lake County NuRF project. A summary of the chemical characteristics of dried alum residual from NuRF pilot studies is given in Table The alum sludge evaluated during this study was generated by chemical coagulation of thousands of gallons of water collected from the Apopka-Beauclair Canal. The generated floc was captured, placed onto a drying bed, and allowed to dewater. After the sludge has dried, chemical characteristics of the sludge were evaluated and compared with Clean Soil Criteria, outlined in Chapter FAC, to assist in identifying disposal options. As seen in Table 2-10, the measured chemical characteristics from the alum residual are substantially less than the applicable Clean Soil Criteria, based upon direct residential exposure which is the most restrictive soil criteria. Based upon this analysis, the dried alum residual easily meets the criteria for use as fill material for daily landfill cover. 43

123 Final Report Chemical Treatment TABLE 2-10 CHEMICAL CHARACTERISTICS OF DRIED ALUM RESIDUAL FROM THE NURF PILOT STUDIES 1 PARAMETER UNITS VALUE CLEAN SOIL CRITERIA 2 (Chap FAC) Aluminum μg/g 51,096 72,000 Antimony μg/g < Barium μg/g < Beryllium μg/g < Cadmium μg/g Calcium μg/g 1,564 None Chromium μg/g Copper μg/g Iron μg/g ,000 Lead μg/g Magnesium μg/g 96.8 None Manganese μg/g ,600 Mercury μg/g < Nickel μg/g Zinc μg/g ,000 NO x μg/g ,000 Total N μg/g 2,054 None SRP μg/g < 1 None Total P μg/g 166 None ph s.u None 1. Residual sample air-dried and screened using an mm sieve 2. Based on residential direct exposure criteria. The alum residual generated during the NuRF project pilot testing was also subjected to the Toxicity Characteristics Leaching Procedure (TCLP) in which a dried residual sample was agitated for 18 hours under acidified conditions at a ph of approximately This procedure resulted in virtually no release of heavy metals from the dried residual samples. Measured concentrations for virtually all heavy metals were less than laboratory detection limits for the evaluated parameters. A summary of the results of the TCLP leachate testing conducted on dried alum residual from the NuRF project is given on Table Soon after initial formation, newly formed alum floc consists of a series of individual crystalline structures which form the newly generated floc. Over time, these crystalline structures begin to combine into larger and larger crystalline structures. During this process, the OH/Al ratio increases approximately 10-fold from 0.3 in the newly formed floc to approximately 3.0 in the aluminum trihydroxide solid phase, commonly called gibbsite. As the crystalline structure becomes larger, the stability of the floc particles increases. The stability of trapped particles and ions within the crystalline structure also increases accordingly. 44

124 Final Report Chemical Treatment TABLE 2-11 RESULTS OF TCLP 1 LEACHATE TESTING ON DRIED ALUM RESIDUAL FROM THE NURF PROJECT PARAMETER UNITS CONCENTRATION Arsenic mg/l < 0.05 Barium mg/l < 1.0 Cadmium mg/l < Chromium mg/l Lead mg/l < 0.05 Mercury mg/l < 0.01 Selenium mg/l < 0.05 Silver mg/l < Toxicity Characteristics Leaching Procedure - sample acidified with acetic acid to ph 4.93 and agitated for 18 hours - residual sample air-dried and screened using an mm sieve A schematic of the aging process for alum floc is given on Figure The aging process summarized on this Figure requires approximately days for completion. Once the floc reaches the final aluminum trihydroxide or gibbsite phase, the structures are extremely stable Aging Process for Alum Sludge under a wide range of ph and redox conditions. Figure 2-14 OH Al = [Al 6 (OH) 12 (H 2 0) 12 ] 6+ [Al 10(OH) 22 (H 2 0) 16 ] 8+ [Al 13 (OH) 30 (H 2 0) 18 ] 9+ OH Al = OH Al = Aluminum trihydroxide solid phase [Al 13 (OH) 30 (H 2 0) 18 ] 9+ [Al 24 (OH) 60 (H 2 0) 24 ] 12+ [Al 54(OH) 144(H 2 0) 36] 18+ [Al n (OH) 3n Conclusions: 1. Aged alum floc is exceptionally stable under a wide range of ph and redox conditions 2. Constituents bound into the floc are inert and have virtually no release potential Figure Process of Aging for Newly Formed Alum Floc. ERD has also conducted laboratory experiments on accumulated alum floc samples collected from Lake Ella, Lake Dot, and Lake Lucerne to evaluate the influence of ph and redox potential 45

125 Final Report Chemical Treatment on the stability of heavy metals in alum treated sediments. Each of these lakes had alum stormwater treatment systems with direct floc discharge which had been operational for 2-3 years. Each of these lakes also received an alum surface application to control sediment phosphorus release. An incubation apparatus was constructed which allows a circulating sediment slurry to be maintained under precisely controlled conditions of ph and redox potential. A schematic and photograph of this apparatus are given on Figure Samples were collected periodically from the sediment slurry to evaluate the solubility of heavy metals within the sediments under various ph and redox conditions. Experiments were conducted at selected ph levels typical of values within the sediments of each lake, as well as redox potentials from highly reduced to highly oxidized. The results of incubation experiments conducted on pre- and post-sediment samples collected from Lake Ella for chromium, copper, lead and zinc are summarized in Figure 2-16 (Harper, 1990). Sediment metal release was found to be substantially less in alum treated samples than observed in pre-treatment samples collected from the lake under a wide range of ph conditions and under redox potentials ranging from highly oxidized to highly reduced. Alum floc is capable of tightly binding heavy metals within the sediments, substantially reducing the potential toxicity of in-place sediments. Similar results were obtained for copper, nickel and lead. As alum floc ages, the freshly precipitated AlOH 3 forms into a series of ringed structures which are extremely stable and which tightly bind phosphorus and heavy metals in a crystalline lattice network. These phosphorus and metal associations, once combined with alum, are apparently inert to changes in ph and redox potential normally observed in a natural lake system. The impact of alum floc on lake sediments has also been evaluated by comparison of pre- and post-treatment sediment pore water concentrations in Lake Ella, Lake Lucerne and Lake Cannon. A comparison of sediment pore water concentrations in Lake Lucerne before and after alum stormwater treatment is given in Table Post-treatment samples reflect approximately four years of operation of the alum stormwater treatment system. Introduction of alum floc into the lake sediments has significantly reduced measured concentrations of total nitrogen, total phosphorus and each of the listed heavy metals. Pore water concentrations of total aluminum have also been reduced as a result of replacing pre-treatment aluminum associations with stable Al(OH) 3 associations. The reduced pore water concentrations indicated in Table 2-12 provide an enhanced environment for sediment-dwelling organisms. 46

126 Final Report Chemical Treatment Figure 2-15 Schematic of Sediment Incubation Apparatus Figure 2-15 Incubation apparatus capable of incubating sediments under a wide range of ph and redox conditions (oxidized to reduced) Figure Sediment Incubation Apparatus used to Evaluate Floc Stability. 47