In-Canal Phosphorus Treatment Study for Barr Lake Project Kick-off - - PowerPoint PPT Presentation

In-Canal Phosphorus Treatment Study for Barr Lake

Project Kick-off Meeting

Harvey H. Harper, Ph.D., P.E. Environmental Research & Design, Inc.

Barr Lake/Milton Reservoir Watershed Association January 28, 2014

Uses for Alum in Lake Management

• 1. Treatment of external inflows
• Alum treatment of stormwater and surface water
• ther flows to remove nutrients, particularly

phosphorus

• 2. Inactivation of internal phosphorus recycling
• Alum addition to sediments to bind available

phosphorus

Characteristics of Alum

• Clear, light green to

yellow solution, depending on Fe content

• Liquid is 48.5% solid

aluminum sulfate by wt.

• Specific gravity = 1.34
• 11.1 lbs/gallon
• Freezing point = 5° F
• Delivered in tanker

loads of 4500 gallons each Alum is made by dissolving aluminum ore (bauxite) in sulfuric acid

History of Alum Usage

Drinking water – Roman Times Wastewater – 1800s Lake surface – 1970 Stormwater – 1986

Alum is used to make many common items, such as:

• pickles
• baseballs
• antacids
• deodorants
• vaccines

Colloidal Runoff Sample After 12 Hours Immediately Following Alum Addition Initial Experiments (1980)

Initial testing evaluated salts of:

• Aluminum
• Iron
• Calcium

Alum was most effective Alum Reacts Quickly to Remove Both Particulate and Dissolved Pollutants

Treatment Effectiveness vs. Aluminum Dose

Treatment efficiency improves with increasing alum dose up to the “optimum” dose at which no significant improvement in effectiveness occurs

Significant Alum Removal Processes

• 1. Removal of suspended solids, algae,

phosphorus, heavy metals and bacteria: Al

+3

+ 6H O

2

Al(OH)

3(ppt)

+ 3H

3

O

+

• 2. Removal of dissolved phosphorus:

Al

+3

+ H

n

PO

4 n-3

AlPO

4(ppt)

+ nH

+

Aluminum Coagulants

Aluminum Sulfate (alum) Aluminum Chloride Poly-Aluminum Chloride Alum/Polymer Blends (floc logs)

Alum Coagulation

Advantages

• Rapid, efficient removal of solids, phosphorus, and bacteria
• Inexpensive – approximately $0.60/gallon
• Low contaminant levels
• Relatively easy to handle and feed
• Does not deteriorate under long-term storage
• Floc is inert and is immune to normal fluctuations in pH and redox
• Floc binds heavy metals in sediments, reducing sediment toxicity

Disadvantage

• May result in lowered pH and elevated levels of Al+3 if improperly

applied

• 1. Treatment of External

Inflows

History of Chemical Stormwater Treatment

Initial research on chemical coagulation conducted in the late 1970s – Evaluated salts of Al, Fe, and Ca Chemical coagulation evaluated for several stormwater retrofit projects in the early 1980s First system constructed at Lake Ella in Tallahassee in 1986 Currently, 60 systems have been designed and constructed in Florida with 4-5 in other states Systems have been designed to treat a wide variety of inflow types

Procedures For Evaluation Of Alum Treatment Feasibility

• 1. Collect representative samples of inflow to be treated
• Include stormwater as well as dry weather baseflow, if present
• Samples should reflect anticipated range of water quality characteristics
• 2. Perform jar testing to evaluate:
• pH response to alum addition
• floc formation rates and settling characteristics
• removal efficiencies for constituents of interest
• 3. Perform hydrologic modeling to:
• evaluate range of flows to be treated
• estimate annual volume to be treated
• establish design parameters for process equipment
• 4. Evaluate floc collection and disposal options
• floc collection may or not be required depending on the receiving water
• floc may be collected in a dedicated settling pond
• collection and disposal to sanitary sewer
• direct inflow into receiving water

Typical Percent Removal Efficiencies for Alum Treated Stormwater Runoff

Parameter Settled Without Alum (24 hrs) Alum Dose (mg Al/liter) 5 7.5 10

Ammonia ~ 0 ~ 0 ~ 0 ~ 0 NOx ~ 0 ~ 0 ~ 0 ~ 0

• Diss. Organic N

20 51 62 65 Particulate N 57 88 94 96 Total N 15 ~ 20 ~ 30 ~ 40

• Diss. Ortho-P

17 96 98 98 Particulate P 61 82 94 95 Total P 45 86 94 96 Turbidity 82 98 99 99 TSS 70 95 97 98 BOD 20 61 63 64 Fecal Coliform 61 96 99 99

• Removal efficiencies for waters with elevated color will be lower

• Surface area = 29

acres (11.7 ha)

• Lake divided into

eastern and western lobes by 6 lane road

• 267 acre

watershed

• Six primary

inflows contribute 95% of annual runoff

• Mean depth = 10

ft (3 m)

• Pre-modification

TP conc. > 100 µg/l

Lake Lucerne – Orlando Southern Gateway

Lake Lucerne (21.0 ac.)

Mechanical components for the Lake Lucerne alum treatment system are housed in an underground vault beneath an elevated expressway

Chemical metering pumps Pump control panels Flow meter control panels

Lake Lucerne following start-up of the alum stormwater treatment system

20 40 60 80 100 120 140 160 180

Total Phosphorus (µg/l)

1990 1991 1992 1993 1994 1995 1996

Date Total Phosphorus

Before Alum Testing / Startup During System Operation During System Operation System Offline

Lake Lucerne

Pre-treatment Water Quality 108 inch Stormsewer Post Treatment Water Quality

Lake Dot – Orlando

5 ac. Lake Receiving Runoff from 305 ac. Urban Watershed

Newspaper Cartoon

Equipment Building Alum Injection Equipment Underground Alum Storage Tank

Lake Howard

Floc Discharge to Lake

Equipment Building Floc Disposal System In-line Floc Trap In-line Floc Trap

Gore Street – Clear Lake

Floc Pumping Equipment Permeable Fabric

Equipment Building Alum Injection Equipment pH Control Equipment In-line Floc Settling Pond

Merritt Ridge

Regional Flood Control Pond used for Floc Collection

Drivable Drainage Diversion Weir

Largo Regional Alum Treatment System

Treated Watershed Area = 1500 acres

Alum Injection Building Canal Flow Diverted Into Box Culvert Flow

Floc Settling Pond Elevated Wooden Boardwalk Floating Dock Paved Walking Path 15 Acre Hardwood Wetland Enhancement Wetland Enhancement

Largo Regional Alum Treatment System Components

Inflow Outflow

• Diss. Ortho-P
• Diss. Organic P

Particulate P Total P

Phophorus Concentration (µg/l)

50 100 150 200 250 Inflow Outflow

Nutrient Removal in the Largo Alum Stormwater Facility

NH3-N NOx

• Diss. Organic N

Particulate N Total N

Nitrogen Concentration (µg/l)

500 1000 1500 2000 2500 Inflow Outflow

Lake Apopka (12,000 ha) Apopka-Beauclair Canal Lock & Dam Lake Beauclair Lake Dora NuRF Site

LCWA Nutrient Reduction Facility (NuRF)

Overview of NuRF Project

Pond 2

Area = 8.2 ac. Depth = 20 ft.

Pond 1

Area = 8.2 ac. Depth = 20 ft.

Lock and Dam Structure Inflow Canal Outflow Canal Apopka- Beauclair Canal Pump and Control Building 6 – 12,000 gal Storage Tanks Dried Floc Storage Area Dewatering Building Storage/Mixing Tank Alum/Air Addition

Specifications

• Treat flows to 300 cfs
• Cost = 7.2 million
• Alum Use = 1.5 – 2.9 million gal/year
• 35,078 gal/day at peak flow
• Treats 89% of annual canal flow
• TP Removal = 10,000 kg/yr
• 10-20,000 ft3 dried floc/yr
• Floc used as soil amendment
• P removal cost = $200/kg

Estimated Annual Discharges Through the Apopka-Beauclair Canal

Condition Annual Canal Discharge (ac-ft/yr) Estimated Annual Mass Load (kg/yr) Total N Total P TSS BOD Existing 1959-2000 54,092 193,972 13,328 2,465,472 339,836 Post Treatment1 54,092 137,002 (-29%) 3,838 (-71%) 591,713 (-76%) 209,781 (-38%)

• 1. Assumes that the system will treat 89% of water on an annual basis

Lake Seminole Bypass Canal and Basin 1 Alum Treatment Systems

Bypass Canal Basin 1 Canal

Pump and Control Building Treated Water Discharges to Floc Settling Trough Pumped Inflow ~ 10 cfs Alum Added Treated Water Discharge

Lake Seminole

In-Lake Floc Settling Area Alum/Air Addition Floc Pumped to Sanitary Pump Station

Seminole Bypass Canal Alum pumping and control bldg. Treatment System Inflow 10 cfs Treated Discharge

Lake Seminole

Cross-section of Treatment System

Floc collection system – discharge to sanitary sewer

25 ft.

Lake Seminole Bypass Canal Treatment System

First system which is totally automated

Bypass Canal Floc Collection Trough

Pumped Inflow Pump/Control Building Water Level Control Weir

Inflow Portion of Floc Collection Trough Floc Collection System

Floc Collection Piping

PLC Pump and System Controller Floc collection Control Valve

Floc Pumped to Sanitary Sewer

Aluminum Solubility

After addition of alum, the concentration of dissolved aluminum is regulated by the pH of the treated water

– Minimum solubility of dissolved aluminum occurs in the pH range of 5.5 – 7.0 – If the pH of the treated water is in this range, dissolved aluminum will be minimal – Diss. Al concentrations generally decrease after treatment

5.7 7.2 Area of solubility < 100 ppb

Solid phase

Aluminum Concentrations in Treated Water

Typical dissolved aluminum concentrations:

– Alum treated water ~ 50 -100 ppb – Most stringent EPA recommendation for most sensitive species in U.S. ~ 87 ppb (cold water fish in Washington) – Drinking water ~ 200 ppb; Milk ~ 700 ppb – Steeped tea ~ 4600 ppb, strong coffee ~ up to 10,000 ppb

Numerous fish and zooplankton bioassays have been conducted on alum treated water with 100% survival in all tests

Alum floc Pond sediments

Alum floc settles onto the pond bottom and begins to consolidate

Anticipated Production of Alum Sludge from Alum Treatment of Stormwater at Various Doses

Alum Dose (mg/l as Al) Sludge Production1 As Percent of Treated Flow Per ac-ft of Runoff Treated 5 0.16 70 ft3 7.5 0.20 87 ft3 10 0.28 122 ft3

• 1. Based on a minimum settling time of 30 days

[Al6(OH)12(H20)12]6+ [Al10(OH)22(H20)16]8+ [Al13(OH)30(H20)18]9+ OH Al = 0.3-2.1

Aging Process for Alum Sludge

Aluminum trihydroxide solid phase (Gibbsite) [Al13(OH)30(H20)18]9+ [Al24(OH)60(H20)24]12+ [Al54(OH)144(H20)36]18+ [Aln(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

OH Al = 2.2-2.7 OH Al = 3.0-3.3

Evaluating Stability of Alum Treated Sediments

Impacts of alum addition to sediments was evaluated using a sediment incubation apparatus Allows sediment slurry to be incubated under a variety of pH and redox conditions Testing was conducted using both pre and post treatment sediments Redox conditions varied from highly reduced to oxidized conditions Redox potential controlled automatically using a relay to inject either air or nitrogen as needed After desired conditions were reached, the sediments were incubated for 7-10 days and a sub- sample collected for lab analysis

0.0 0.3 0.5 0.8 1.0 5.0 6.5 5.0 6.0 7.0 pH Cadmium Release (µg/g Dry Wt.)

• 200 mV

0 mV 200 mV 400 mV

Pre Post

Lake Ella Sediment Metal Release in Treated an Untreated Sediments

0.00 0.25 0.50 0.75 1.00 1.25 5.0 6.5 5.0 6.0 7.0 pH Chromium Release (µg/g Dry Wt.)

• 200 mV

0 mV 200 mV 400 mV 0.00 0.25 0.50 0.75 1.00 1.25 5.0 6.5 5.0 6.0 7.0 pH Copper Release (µg/g Dry Wt.)

• 200 mV

0 mV 200 mV 400 mV 2 4 6 8 10 12 14 5.0 6.5 5.0 6.0 7.0 pH Lead Release (µg/g Dry Wt.)

• 200 mV

0 mV 200 mV 400 mV

Release (µg/g dry wt.)

Lead

Release (µg/g dry wt.)

Copper Chromium Cadmium

Release (µg/g dry wt.) Release (µg/g dry wt.)

Post Pre Post Pre Pre Post

5 10 15 20 25 5.0 6.5 5.0 6.0 7.0 pH Aluminum Release (µg/g Dry Wt.)

• 200 mV

0 mV 200 mV 400 mV 0.00 0.50 1.00 1.50 2.00 5.0 6.5 5.0 6.0 7.0 pH Zinc Release (µg/g Dry Wt.)

• 200 mV

0 mV 200 mV 400 mV 0.0 0.2 0.4 0.6 0.8 1.0 1.2 5.0 6.5 5.0 6.0 7.0 pH Nickel Release (µg/g Dry Wt.)

• 200 mV

0 mV 200 mV 400 mV

Pre Post

Lake Ella Sediment Metal Release in Treated an Untreated Sediments

Release (µg/g dry wt.)

Aluminum Zinc Nickel

Release (µg/g dry wt.) Release (µg/g dry wt.)

Post Pre Post Pre

Sediment Pore Water Concentrations in Lake Lucerne

Parameter Units Pre-Treatment (12/92) Post Treatment (3/96) Percent Change (%) Total N µg/l 9978 9765

• 2

Total Al µg/l 417 242

• 42

Total Cu µg/l 21 8

• 62

Total Fe µg/l 1389 467

• 66

Total Ni µg/l 17 2

• 88

Total Mn µg/l 314 77

• 75

Total Zn µg/l 80 30

• 63

Comparison Of Dominant Pre-treatment And Post-treatment Macroinvertebrate Assemblage At Site 1 In Lake Mizell

TAXA 1/28/97 1/29/98 1/27/99 1/31/00 MEAN (#/m2) % MEAN (#/m2) % MEAN (#/m2) % MEAN (#/m2) % Chaoborus punctipennis 4664 77.9 253 41.7 30 2.0 4502 66.3 Chironomus sp. 647 10.8 74 12.2 1095 73.2 119 1.7 Limnodrilus hoffmeisteri 396 6.6

• 74

4.9 592 8.7 Procladius bellus 15 0.2 148 24.4

• 252

3.7 Tanytarsus sp. 30 0.5 58 9.6

• Ablabesmyia rhamphe group
• 44

7.2

• Cladopelma sp.
• 15

2.2

• 30

0.4 Hyalella azteca 57 0.9 15 2.2

• 296

4.4 Dero Nivea

• 237

15.8 74 1.1 Dero Trifida

• 15

1.0 30 0.4

• Unid. Ceratopogonidae
• 30

2.0 15 0.2 Thienemanniella sp.

• 15

1.0

• Glyptotendipes paripes
• 726

10.7 Pristina sp.

• 59

0.9 Cryptochironomus sp.

• 59

0.9 Ablabesmyia peleensis

• 15

0.2 Chaetogastor diaphanus

• 15

0.2

Lake Mizell received alum floc from treatment of runoff and a whole lake alum treatment

Alum Floc Drying Process

After 4-7 days After 30 days

Once completely dried, the floc forms into a rock hard material that will not re- dissolve

Floc after initial water decanting

Floc color is a function

• f the materials

removed from the treated water

Chemical Characteristics of Dried Alum Residual from the NuRF Pilot Studies1

Parameter Units Value Clean Soil Criteria2 (Chap. 62-777 FAC)

Aluminum μg/g 51,096 72,000 Antimony μg/g < 6.3 26 Barium μg/g < 21 110 Beryllium μg/g < 0.53 120 Cadmium μg/g 0.5 75 Calcium μg/g 1,564 None Chromium μg/g 65.0 210 Copper μg/g 31.6 110 Iron μg/g 764 23,000 Lead μg/g 0.7 400 Magnesium μg/g 96.8 None Manganese μg/g 12.3 1,600 Mercury μg/g < 0.091 3.4 Nickel μg/g 2.3 110 Zinc μg/g 50.6 23,000 NOx μg/g 0.773 120,000 Total N μg/g 2,054 None SRP μg/g < 1 None Total P μg/g 166 None pH s.u. 6.17 None

Alum residual easily meets the clean soil criteria

Characteristics of TCLP Leachate Testing on Dried Alum Residual from NuRF

Parameter Units Measured Value EPA Regulated Level Class III Criteria (Ch. 62-302 FAC) Arsenic µg/l < 1 < 5000 < 50 Cadmium µg/l < 0.3 < 1000 < 0.3 Chromium µg/l 11 < 5000 < 86 Lead µg/l < 2 < 5000 < 3 Mercury µg/l < 0.01 < 200 < 0.01 Selenium µg/l < 5 < 1000 < 5 Silver µg/l < 007 < 5000 < 0.07 Leaching testing conducted under acid conditions at pH of 4.93

Comparison of Life Cycle Cost Per Mass Pollutant Removed for Typical Stormwater Retrofit Projects*

Project 20–Year Life Cycle Cost ($) Cost per Mass Pollutant Removed ($/kg) TP TN TSS

Alum Treatment Largo Regional STF Lake Maggiore STF Gore Street Outfall STF East Lake Outfall TF LCWA NuRF Facility Mean 2,044,780 4,086,060 1,825,280 1,223,600 34,254,861 253 200 87 135 198 164 65 71 12 17 30 31 4 2 1 1 2 2 Wet Detention Melburne Blvd. STF Clear Lake Ponds STF Mean 1,069,000 1,091,600 371 658 494 125 237 172 2 22 2

* Does not consider land cost

• 2. Inactivation of Internal

Sediment Recycling

Typical Zonation in a Lake

• Anoxic sediments release P, Fe, Mn, ammonia and other ions
• Alum floc binds with P, forming an insoluble, inert precipitate

which eliminates sediment P release

Ferrous-Ferric System

• The retention of P in the sediments of many lakes is

regulated by bonding mechanisms with iron

• Under aerobic conditions Fe-P bonds are insoluble but

become soluble under anoxic conditions (< 0 mV)

Fe+++ + e- < = > Fe++

• This process is preceded by nitrate reduction and

release of alkalinity, CO2 and NH3

Ferric iron Oxidized State Insoluble Ferrous iron Reduced State Soluble

Floc initially settles onto the surface of the sediments

Floc Settling in a Shallow Lake

Alum floc layer Consolidated organic sediments Unconsolidated organic sediments (muck) Floc migrates downward over time Alum floc layer

• Alum floc initially settles onto

the top of the loose surficial layer

• Floc migrates downward over

time into unconsolidated sediment layer

• If the alum treated sediment

re-suspends as a result of wind

• r boating activities, then it will

quickly settle back

• This will have no impact on the

effectiveness since the sediment P will be adsorbed onto the floc

• Since the alum floc still

maintains effectiveness, floc re- suspension may adsorb and remove additional P from the water column

ρ ~ 1.05 ρ ~ 1.02 ρ ~ 1.1-1.2

Project Objectives

Identify and evaluate passive and active treatment options that will reduce inflow TP by 75%.

• Alum treatment
• Infiltration
• Wetland

Provide cost estimates for each in-canal treatment

• ption

Provide a comprehensive report that summarizes Tasks 1 and 2 and provides recommendations

• n the next steps to be

taken in implementing an in-canal phosphorus treatment system

Imagery Date: October 2012

Barr Lake Inflow Characteristics

(11/1/06 – 10/31/09)

50 100 150 200 250 300 350 400 Number of Days Interval (cfs)

150 60

Inflow Water Quality Characteristics

2006 2007 2008 2009 2010 2011 2012 2013

pH

7.0 7.5 8.0 8.5 9.0 9.5 2006 2007 2008 2009 2010 2011 2012 2013

Alkalinity (mg/l)

50 100 150 200 250 300 350

• Typical values between 7.7 - 8.7
• Alum treatment will reduce

inflow pH to 6 - 7 range

• Typical values 75 - 200 mg/L
• Alkalinity sufficient for alum

treatment

Inflow Water Quality Characteristics

• Alum treatment adds ~ 3 mg/L

sulfate for every 1 mg/L Al added

• A typical alum dose of 10 mg

Al/L will add 30 mg/L sulfate

• Temperature impacts the rate of

floc formation

• Lab jar testing will be conducted
• n refrigerated samples to

simulate low temps

2006 2007 2008 2009 2010 2011 2012 2013

Sulfate (mg/l)

100 200 300 400 500 600

Conductiviy (µmho/cm)

2006 2007 2008 2009 2010 2011 2012 2013

Temperature (°C)

5 10 15 20 25 30

Inflow Water Quality Characteristics

• Total P is highly variable
• Majority of total P is SRP
• SRP is easily removed by alum,

but produces a slow settling floc

• Total N is also highly variable
• Alum has no impact on NOx, but

does remove portions of TKN

2006 2007 2008 2009 2010 2011 2012 2013

Nitrogen (mg/l)

2 4 6 8 10 12 14 16 18 NOx TKN Total N 2006 2007 2008 2009 2010 2011 2012 2013

Phosphorus (mg/l)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 SRP Total P

Preliminary Treatment Evaluation

• Mass loadings of TP were calculated by multiplying the TP conc. times the flow
• Analysis conducted to estimate annual mass removal assuming a TP removal of

80% by alum

• Simulations were conducted to determine the TP removal by treating inflows up

an assumed discharge rate until a removal of 75% was achieved

100 200 300 400 500 600 700 800 900 1,000 01/01/06 01/01/07 01/01/08 01/01/09 Phosphorus (kg/day) Treated Load (kg/day) Untreated (kg/day)

• A 75% annual mass removal was obtained by treating flows up to 139 cfs

Wetland Settling Option

Baker Creek Inflow Constructed Wetland Treatment Area Flow Diversion

Option evaluated to divert alum treated water into the wetland treatment area However, several concerns led to elimination of this

• ption

– Wetland is a permitted mitigation site – Wetland was shallow, and concern raised over filling of wetland with floc – Difficulty in floc removal – Floc removal would damage wetland – Concerns over wetland area negatively impacting treatment efficiency

Questions?