Page 1

Editor’s Office and Advertiser Information:

Florida Water Resources Journal 1402 Emerald Lakes Drive Clermont, FL 34711 Phone: 352-241-6006 • Fax: 352-241-6007 Email: Editorial, Display and Classified Advertising,

Business Office: P.O. Box 653, Venice, FL 34284-0653 Web: General Manager: Editor: Graphic Design Manager: Mailing Coordinator:

Michael Delaney Rick Harmon Patrick Delaney Buena Vista Publishing

Published by BUENA VISTA PUBLISHING for Florida Water Resources Journal, Inc. President: Richard Anderson (FSAWWA) Peace River/Manasota Regional Water Supply Authority Vice President: Lisa Prieto (FWEA) Prieto Environmental LLC Treasurer: Rim Bishop (FWPCOA) Seacoast Utility Authority Secretary: Holly Hanson (At Large) ILEX Services Inc., Orlando

Moving? The Post Office will not forward your magazine. Do not count on getting the Journal unless you notify us directly of address changes by the 15th of the month preceding the month of issue. Please do not telephone address changes. Email changes to, fax to 352-241-6007, or mail to Florida Water Resources Journal, 1402 Emerald Lakes Drive, Clermont, FL 34711

Membership Questions FSAWWA: Casey Cumiskey – 407-957-8447 or FWEA: Karen Wallace, Executive Manager – 407-574-3318 FWPCOA: Darin Bishop – 561-840-0340

Training Questions FSAWWA: Donna Metherall – 407-957-8443 or FWPCOA: Shirley Reaves – 321-383-9690

For Other Information DEP Operator Certification: Ron McCulley – 850-245-7500 FSAWWA: Peggy Guingona – 407-957-8448 Florida Water Resources Conference: 407-363-7751 FWPCOA Operators Helping Operators: John Lang – 772-559-0722, e-mail – FWEA: Karen Wallace, Executive Manager – 407-574-3318

Websites Florida Water Resources Journal: FWPCOA: FSAWWA: FWEA: and Florida Water Resources Conference: Throughout this issue trademark names are used. Rather than place a trademark symbol in every occurrence of a trademarked name, we state we are using the names only in an editorial fashion, and to the benefit of the trademark owner, with no intention of infringement of the trademark. None of the material in this publication necessarily reflects the opinions of the sponsoring organizations. All correspondence received is the property of the Florida Water Resources Journal and is subject to editing. Names are withheld in published letters only for extraordinary reasons. Authors agree to indemnify, defend and hold harmless the Florida Water Resources Journal Inc. (FWRJ), its officers, affiliates, directors, advisors, members, representatives, and agents from any and all losses, expenses, third-party claims, liability, damages and costs (including, but not limited to, attorneys’ fees) arising from authors’ infringement of any intellectual property, copyright or trademark, or other right of any person, as applicable under the laws of the State of Florida.

News and Features 9 10 23 33 51 52

Bonita Springs Utilities Elects Officers Drop Savers Poster Contest Winners Announced Low Concentrations of Silver Can Foil Wastewater Treatment Bonita Springs Utilities Announces Water Conservation Poster Contest Student Winners News Beat WEF HQ Newsletter: WEFTEC® Opening General Session Speaker Highlights How to Be a Water Sector Hero 61 Generators Enhance Utility’s Hurricane Preparedness

Technical Articles 4 Groundwater Nanofiltration Plant Addresses Color and Disinfection Byproducts for Flagler County— Phillip J. Locke, Eric A. Smith, and Mark Ralph 16 Dialing Down Disinfection Byproducts With Chlorine Dioxide Pre-Oxidation—Lance Littrell, Bryan Gongre, Patrick Flynn, Domenic Gentilucci, Steve Romano, Rhea Dorris, and Gina Parra 36 Can Machine Learning and Geostatistics Overcome Lack of Data in Assessing Recovery of Water Levels and Ecological Conditions at Unmonitored Wetlands and Lakes?—Dan Schmutz, Stephanie K. Garvis, Danny Goodding, and Christopher Shea

Education and Training 9 13 19 24 25 26 27 28 29 30 39 43

Florida Water Resources Conference CEU Challenge FWPCOA Training Calendar FSAWWA Fall Conference Overview FSAWWA Fall Conference Exhibits FSAWWA Fall Conference Poker Night and Happy Hour FSAWWA Fall Conference Golf Tournament FSAWWA Fall Conference Competitions FSAWWA Water Distribution System Awards FSAWWA Water Conservation Awards TREEO Center Training FWPCOA Short School

Columns 12 14 32 34 50

Let’s Talk Safety FSAWWA Speaking Out—Bill Young C Factor—Mike Darrow FWEA Focus—Kristiana S. Dragash Contractors Roundup: The Owner’s Role in a Construction Project—Lauren C. Atwell 54 FWEA Committee Corner: Infiltration and Inflow From Cradle to Grave—Jamison Tondreault

Departments 55 56 59 62

New Products Service Directories Classifieds Display Advertiser Index

Volume 69

ON THE COVER: Two Florida Section AWWA teams, Palm Beach Coast Water Buoys and FWPCOA Region #9, came in second and third, respectively, in the Top Ops competition held at ACE18 in June in Las Vegas. The Palm Beach team members, pictured in the top photo (in the middle left to right) are Fred Greiner, Peter Roussell, Jim Hogan (team coach), and Tom Martens. The FWPCOA team members (in the bottom photo in the middle left to right) are Steve Harrison, Frank Kelsey, Kameron Van Kleeck, and Glenn Whitcomb. Flanking both teams (at the time the photos were taken) are (far left) David Rager, AWWA president-elect, and (far right) Brenda Lennox, AWWA president.

August 2018

Number 8

Florida Water Resources Journal, USPS 069-770, ISSN 0896-1794, is published monthly by Florida Water Resources Journal, Inc., 1402 Emerald Lakes Drive, Clermont, FL 34711, on behalf of the Florida Water & Pollution Control Operator’s Association, Inc.; Florida Section, American Water Works Association; and the Florida Water Environment Association. Members of all three associations receive the publication as a service of their association; $6 of membership dues support the Journal. Subscriptions are otherwise available within the U.S. for $24 per year. Periodicals postage paid at Clermont, FL and additional offices.

POSTMASTER: send address changes to Florida Water Resources Journal, 1402 Emerald Lakes Drive, Clermont, FL 34711

Florida Water Resources Journal • August 2018



Groundwater Nanofiltration Plant Addresses Color and Disinfection Byproducts for Flagler County Phillip J. Locke, Eric A. Smith, and Mark Ralph lagler County (county) owns and operates the Plantation Bay Water Treatment Plant (PBWTP), which was constructed in the 1980s. The PBWTP has a design capacity of 756,000 gal per day (gpd) and utilizes four groundwater wells to produce average day and maximum day flows of 232,000 gpd and 377,000 gpd, respectively. While all of the wells are moderately high in total hardness (~325 mg/L as calcium carbonate [CaCO3]), two of the wells also yield water with high color that frequently ap-


proaches 90 color units. The lime softening treatment process utilized at the PBWTP is ineffective in removing the color, and this condition has resulted in frequent complaints from customers and the inability to use these wells. Additionally, the PBWTP has had past exceedances of the U.S. Environmental Protection Agency (EPA) disinfection byproduct (DBP) maximum contaminant levels (MCL) of 80 µg/L for total trihalomethanes (TTHMs). Both the color and DBP formation issues stem from natural organic

Table 1. Weighted Treatment Decision Matrix

Phillip J. Locke, P.E., is senior project manager with McKim & Creed Inc. in Clearwater. Eric A. Smith, P.E., is a project engineer and Mark Ralph, P.E., is senior project manager with McKim & Creed Inc. in Daytona Beach.

matter (NOM) in the wells, resulting in total organic carbon (TOC) levels as high as 25 mg/L. An alternatives evaluation was performed to determine the best treatment solution to address the water quality issues. A weighted treatment (1 = lowest ranking, 5 = highest ranking) decision matrix was developed for these potential alternatives, as shown in Table 1. Based on results from the evaluation, nanofiltration (NF) was selected for implementation to remove organics, color, and hardness, and to reduce DBP formation potential. Concurrent with preliminary and final design, a pilot test was conducted using the well with the highest organics and color. The county is moving forward with replacing the current treatment process with a lowpressure NF treatment system, as NF is ideally suited for the removal of dissolved constituents, such as TOC and hardness. This article presents the findings from the pilot test and discusses the design of the new NF treatment system.

Pilot Study The PBWTP has four production wells that have varying high levels of iron, ammonia, sulfides, color, TOC, and hardness. Groundwater quality data from the wells is shown in Table 2. Pilot testing was performed to confirm that the proposed treatment process will produce a quality effluent that meets or surpasses treatment goals of the new water treatment project. Based on the groundwater quality presented in Table 2, the pilot study aimed to reduce the following groundwater constituents: S Iron – High levels of iron in the source water Continued on page 6


August 2018 • Florida Water Resources Journal

Continued from page 4 contribute to color within the water and are likely the source of the majority of customer complaints received by the county. S Hardness – With an average hardness of 338 mg/L as CaCO3, the PBWTP source water is considered very hard (>180 mg/L as CaCO3), according to the U.S. Geological Survey (USGS) hardness classification. S Color – In groundwater, color may be attributed to a variety of sources, including metallic ions, organic acids, or dissolved plant materials. Color in the water has been the most common complaint from customers. S TOC – This is a measure of organic matter in water and becomes a major concern when the source water is chlorinated, potentially forming DBPs such as trihalomethanes (THM) and haloacetic acids (HAA5). A pilot unit was designed to address these constituents and produce high-quality potable water. The pilot study simulated the performance of the components of a full-scale system. The primary objectives of the pilot study were to: 1. Demonstrate the filter media operating parameters, including: a. Iron removal efficiency b. Filter head loss as a function of run time c. Approximate filter run length 2. Determine the chemical feed rates to meet the water’s oxidant demand to the extent possible. 3. Evaluate membrane nanofiltration operating parameters, including: a. Ability to remove TOC with a goal less than 1 mg/L b. Hardness removal efficiency c. Comparison of NF membrane performance with Dow® ROSA NF membrane model simulation

d. Transmembrane pressure as a function of run time e. Confirm that antiscalants chemical feed rates recommended by manufacturer avoid fouling the membrane f. Approximate run length before clean-inplace (CIP) is required 4. Demonstrate that the proposed equipment can meet the following water quality objectives: a. TOC below 1 mg/L and near detection limit b. Filter effluent iron below MCL and NF permeate iron near detection limit c. Manganese <0.05 mg/L d. Total Hardness < 150 mg/L e. Total Sulfides < 0.01 mg/L f. Ammonia, nitrogen converted to monochloramine It was determined that the well with the highest color and organics (Well No. 4) would be used for pilot testing. This well was selected to ensure a conservative design approach. Additionally, due to the high levels of color and organics, Well No. 4 was not being used for production purposes and, therefore, provided access for the pilot trailer. A blend of the various wells was considered; however, the existing piping and valving infrastructure limited the blending opportunities. The pilot unit consisted of the following treatment schemes: Filtration A 3-ft-diameter by approximately 7-ft-tall filter column was utilized. Its main purpose was to remove iron and other suspended solids. To aid in the removal of iron, an injection point was included upstream of the filter. Prior to filtration, a small dosage of sodium hypochlorite was added to precipitate iron for downstream

Table 2. Groundwater Quality Data (May 2012)

filtration. It should be noted the low dosage was used to form monochloramines and to limit the formation of DBPs. The filter was set up to monitor differential pressure across the filter, and valves were provided so that filter backwash could be performed at predetermined differentials. Nanofiltration System The NF was the main treatment process used to decrease the concentration of iron, organics (DBP precursors), color, and hardness in the filter effluent. Upstream of the NF system, sodium metabisulfite was added for dechlorination, and antiscalant was added to reduce the potential for scaling, especially at the tail end of the second-stage membrane elements. The NF system also included a 5-micron cartridge filter for filtering out any particulates in the filter effluent. The system also utilized a high-pressure pump to push water through the NF membranes. A twostage configuration was utilized with a 2/1 array. The first stage incorporated Dow NF90-4040 membrane elements, while the second stage utilized the “more open” Dow NF270-4040 membrane elements. ChemScan® A Chemscan spectroscopic analyzer was also utilized with this pilot to provide online testing of color, TOC, and iron levels in the raw water and permeate water.

Pilot Study Performance Iron Removal Based on laboratory results, the iron in the filter effluent averaged approximately Fe = 0.25 mg/L entering the NF membrane system, while the NF permeate iron was below the detection limit of approximately Fe < 0.03 mg/L. Thus, the NF average approximate removal of the NF influent iron was at least 88 and 89 percent, based on the laboratory and field results, respectively. Total Organic Carbon Removal The NF membrane skid provided excellent removal of organics based on laboratory results, lowering the approximate concentration of TOC = 22 mg/L in the raw water and filter effluent to below the detection limit of approximately TOC < 0.57 mg/L in the NF permeate. Calcium, Magnesium, and Hardness Treatment The raw water hardness (avg = 310 mg/L as CaCO3) was effectively decreased with the twostage NF membrane system to an average permeate of 105 mg/L as CaCO3. The hardness of 105 mg/L as CaCO3 is classified as a moderately Continued on page 8


August 2018 • Florida Water Resources Journal

Continued on page 8 hard water by USGS; thus, the hardness was reduced by approximately 66 percent. Totally Dissolved Solids The two-stage NF membrane was demonstrated to effectively decrease the totally dissolved solids (TDS). The raw water averaged TDS of approximately TDS = 390 mg/L, while the TDS in the NF permeate averaged approximately TDS = 160 mg/L. The TDS was lowered by approximately 61 percent. Conductivity The raw water conductivity, which as measured by the laboratory averaged approximately 600 mho/cm, was lowered by the two-stage NF system to an average in the NF permeate of approximately 260 mho/cm; thus, the NF membranes decreased the conductivity by approximately 57 percent. Color The average raw water color based on the Chemscan was approximately 33.9 = PlatinumCobalt (Pt-Co) color units, whereas the color wheel produced raw water readings of approximately 30 color units. Because all of the color wheel readings for the NF permeate were “0” color units, the Chemscan unit, with its multiwavelength analyzer, could discern minor differences in color for the NF permeate effluent. The average, including these points, was approximately 1.1 Pt-Co color units with a range of approximately <1.0 – 1.8 Pt-Co color units.

Pilot Study Summary The pilot study demonstrated that a twostage NF membrane system provided excellent organic removal, while decreasing the hardness to acceptable levels. The organics removal from the NF process lowered the Well No. 4 TOC from approximately 22 mg/L to below detectable limits. The DBP formation potential testing was performed on the NF permeate and indicates that the county will now be able to consistently meet DBP regulations. The results of the pilot study are included in Table 3. It is noted that toward the end of pilot testing, permeate flow from the second stage was significantly reduced and scaling was indicated.

Both an alkaline and acid CIP were performed to restore membrane performance; however, the membranes never fully recovered. It appears that the reduced performance after the cleaning in place (CIP) was the result of iron fouling (not targeted with the CIP) that occurred throughout the pilot operation. The iron fouling likely occurred due to inconsistent operation resulting from Hurricane Irma. Additionally, it's noted that, in order to facilitate daily pilot operation, county staff was utilized to start/stop the unit. Several instances occurred in which chemical levels were not checked prior to starting the system, resulting in a chemical running out while the rest of the system ran. This likely contributed to iron fouling.

Booster Pump Station An inline booster pump station will be constructed and will include three pumps to provide ample feed pressure to the pressurized filters so that a minimum of 20 pounds per sq in. (psi) is available at the suction side of the NF feed pumps.

Design Considerations

Softening and Color Removal Via a membrane separation process, the NF membranes will treat the filtered water to meet all drinking water standards. The NF will remove the color and will allow some hardness and alkalinity to pass through the process, thereby reducing costs associated with post-treatment stabilization. The NF skids will include 5-micron cartridge filters to further protect the membranes, and high-pressure feed pumps provide approximately 90 psi at the membrane inlet. A CIP skid will provide for periodic cleaning using high-pH solutions for biofouling and lowering pH solutions to remove scaling.

Based on the pilot testing and the design team’s experience with other facilities having similar water quality, the treatment processes will include oxidation, pressurized multimedia filtration, dechlorination, pH adjustment, antiscalant, cartridge filtration, membrane softening, permeate stabilization, and disinfection. The system will include a two-stage NF system with an overall recovery of approximately 80 percent. A partial NF bypass stream is planned and will be used to add pH and alkalinity to the permeate water, resulting in lower chemical usage and operational costs. Since the chloride levels of the water supply are so low (~25 mg/L), the concentrate from the NF system will be beneficially used for reuse water irrigation with no damage to plants, grass, and other landscaping. The concentrate will be blended with the reclaimed water to supplement the county’s reclaimed water system that is used to irrigate the Plantation Bay Country Club. This approach to concentrate reuse also provides the benefit of eliminating the need for deep well injection or surface water discharge. The new treatment facility will include a membrane softening treatment process designed to improve water quality by removing hardness, organics, color, and other contaminants. The NF process will include three skids capable of producing a combined 756,000 gpd of finished water. The process will be expandable to 1 mil gal per day (mgd) as demands increase. A brief description of the major components of the reverse osmosis (RO) process is provided.

Table 3. Pilot Study Results


August 2018 • Florida Water Resources Journal

Filtration Three vertical pressure filters will be installed upstream of the RO skids as a means of pretreating the raw water. The filters will mainly serve to remove iron so as to mitigate iron fouling, which typically occurs on the lead elements of the first NF stage. Sodium hypochlorite will be used as an oxidant to aid in iron removal.

Chemical Treatment As previously mentioned, chemicals will be added throughout the treatment process as a means of enhancing the overall treatment and endwater quality. The chemicals that were used for the new treatment process include the following: S Oxidant – Sodium hypochlorite will be added to the raw water line prior to the vertical pressure filters to aid in iron removal. S Dechlorination –The filtered effluent will be dechlorinated, using sodium metabisulfite prior to introduction into the NF membranes. S Antiscalant – To protect the membranes from scaling, an antiscalant will be added to the filtered effluent ahead of the RO treatment process. S Caustic – Sodium hydroxide will be added to the RO permeate stream to increase the pH and alkalinity. S Additional space will be provided for a future chemical as needed. The project also includes a new prefabricated metal building, replacement of the existing filter backwash pumps, yard piping modifications, site improvements, and instrumentation and electrical improvements. Construction and commissioning are scheduled for completion by mid-2020. S

Bonita Springs Utilities Elects Officers





The Bonita Springs Utilities Inc. board of directors has elected Robert Bachman as president, Brian Farrar as vice president, and Ben Nelson Jr. as secretary. Mike Malloy was re-elected treasurer for a second term. Robert Bachman is owner and president of WBG SW Florida Inc. A board member since 2000, he has served previous terms as president, vice president, and treasurer. Brian Farrar joined the board in 2016 and is president and managing member of BCF Management Group LLC. He serves on the Lee County Mosquito Control District board of commissioners and is vice chair of the CREW Land & Water Trust board of directors. A Bonita Springs resident since 1960, Ben Nelson Jr. previously served on the board from 1990 to 2001 and rejoined in 2016. He has served previous terms as president and vice president. He served for 16 years on the Bonita Springs City Council, including two terms as mayor, and has owned and operated Nelson Marine Construction for more than 35 years. Mike Malloy has served on the board since 2012. He recently retired as a vice president of customer service with Mach Energy. A Bonita Springs resident since 2003, he has 40 years of experience in the utility industry and served in management positions with several utility companies. The BSU members elect nine directors to the board to govern the utility. Other board members are Paul J. Attwood, Frank W. Liles Jr., Vincent J. Marchesani, Robert H. Sharkey, and James Strecansky. Bonita Springs Utilities Inc. is a not-forprofit water and wastewater utility cooperative founded by local citizens in 1970. The memberowned utility provides service in the City of Bonita Springs, the Village of Estero, and unincorporated South Lee County. S

Florida Water Resources Journal â&#x20AC;˘ August 2018


Drop Savers Poster Contest Winners Announced Melissa Velez Every year the Florida Section of the American Water Works Association (FSAWWA) sponsors the "Drop Savers” Water Conservation Poster Contest. Students from Kindergarten to 12th grade are encouraged to create a poster

DIVISION 1 – FIRST PLACE Gainesville Regional Utilities Harper Fitzpatrick

depicting a water conservation idea, in slogan form, drawing form, or both. The contest allows students to promote water awareness and the importance of water conservation in their daily routines. Posters are designated under one of the following categories:

DIVISION 1 – SECOND PLACE Citrus County Utilities Keegan Woodhouse

DIVISION 2 – FIRST PLACE Miami-Dade Water and Sewer Giana Pantaleon

DIVISION 3 – FIRST PLACE Miami-Dade Water and Sewer Lina Tamer Elattar


August 2018 • Florida Water Resources Journal

Division 1 - Kindergarten and First Grade Division 2 - Second and Third Grade Division 3 - Fourth and Fifth Grade Division 4 - Middle School: Grades Six, Seven, and Eight Division 5 - High School: Grades Nine, Ten, Eleven, and Twelve

DIVISION 1 – THIRD PLACE City of Leesburg Talon Hardy

DIVISION 2 – SECOND PLACE Town of Davie Anabia Anwar

DIVISION 3 – SECOND PLACE Bonita Springs Utilities Emely Meneses

DIVISION 2 – THIRD PLACE City of Casselberry Mikayla Thomas

DIVISION 3 – THIRD PLACE Gainesville Regional Utilities Cecilia Duda

S Poster are drawn on 8 ½-in. x 11-in. white paper (horizontally or vertically) S Each poster must portray a water conservation idea in a slogan, drawing, or both. Students may use crayons, paint, color pencils, or markers. No highlighters, photos, or computer graphics are permitted. S Students must work on posters individually, otherwise posters will be disqualified. S Only original artwork will be accepted (i.e., no trademarked or copyrighted materials). The responsibility of the Drop Savers Committee is to invite and provide each water utility in Florida with the guidelines for running their own poster contest. Once water utilities select their winners, they send the first-place winner’s poster to the Drop Savers Committee,

where they will participate in the state competition. This year, there were over 100 posters from 29 water utilities that participated in the contest. The prizes for this year included: S First-Place Winners: • $100 Amazon gift card • Plaque displaying the poster • $40 gift card for a pizza party • Calendar displaying the poster • Note cards displaying the poster • Water conservation kit • Tote bag • Certificate S Second-Place Winners: • $75 Amazon gift card • Calendar displaying the poster • Note cards displaying the poster


DIVISION 4 – SECOND PLACE City of Margate Casey Cortez


DIVISION 5 – SECOND PLACE Hillsborough County Charlotte Yang

• Water conservation kit • Tote bag • Certificate S Third-Place Winners: • $50 Amazon gift card • Calendar displaying the poster • Note cards displaying the poster • Water conservation kit • Tote bag • Certificate The winning Drop Savers posters are pictured here. Melissa Velez, P.E., LEED AP, is an engineering manager at Black & Veatch in Coral Springs. S


DIVISION 5 – THIRD PLACE Englewood Water District Vivian Lin Florida Water Resources Journal • August 2018


LET’S TALK SAFETY This column addresses safety issues of interest to water and wastewater personnel, and will appear monthly in the magazine. The Journal is also interested in receiving any articles on the subject of safety that it can share with readers in the “Spotlight on Safety” column.

Know What’s Below: Call 811 Before You Dig ith more than 20 million miles of underground utilities in the United States, dangers abound for uninformed diggers. According to the Common Ground Alliance, a national association of underground utility providers, someone in the U.S. damages an underground line once every six minutes. Between 1988 and 2016, 1,815 pipeline incidents caused by unauthorized excavations resulted in 193 deaths, 757 injuries, and nearly $545 million in property damage from fires, explosions, and electrocutions.


All of these instances involved someone digging into underground utilities. Unfortunately, these types of incidents occur thousands of times every year because excavators or homeowners did not call their neighborhood locating service, such as Dig Alert or One Call, ahead of time. Remember that it’s becoming more commonplace for all utilities to be laid in the same trench, so if you’re looking for your water pipes, you may also find gas, electric, and communications lines.

Can You Dig It? Calling 811 is Free and Easy It’s easy to avoid digging into other utility lines. All it takes is a call to 811 from anywhere in the U.S., and you will automatically be connected to your local underground service operator. The name may change from community to community (in Florida, it’s Sunshine 811), but its function is the same: to protect you, your coworkers, and the public. It is imperative that this call be made before beginning any excavation. It’s important, especially for utilities, to use this service because as-built maps and charts are often inaccurate and outdated.

The Five Critical Steps to Safe Digging Before any excavation, the following should always be done: 1. Survey and mark – Survey the proposed excavation areas and mark the dig sites in white paint or chalk. 2. Call before you dig – Call 811 and talk to your local utility locator service. 3. Wait the required time – Allow two working days to have the lines located and marked. 4. Respect the marks – Maintain the marks and follow them when digging. 5. Dig with care – Hand-excavate within 24 inches of each side of the lines. If you hit an underground utility line, you could be hurt or killed. You may also be liable to the other utilities for costly damages and lost service.

Resources Available For more information about specific requirements by state, check out the Common Ground Alliance website at S

The 2017 Let's Talk Safety is available from AWWA; visit or call 800.926.7337. Get 40 percent off the list price or 10 percent off the member price by using promo code SAFETY17. The code is good for the 2017 Let's Talk Safety book, dual disc set, and book + CD set.


August 2018 • Florida Water Resources Journal

Operators: Take the CEU Challenge! Members of the Florida Water and Pollution Control Operators Association (FWPCOA) may earn continuing education units through the CEU Challenge! Answer the questions published on this page, based on the technical articles in this month’s issue. Circle the letter of each correct answer. There is only one correct answer to each question! Answer 80 percent of the questions on any article correctly to earn 0.1 CEU for your license. Retests are available. This month’s editorial theme is Disinfection and Water Quality. Look above each set of questions to see if it is for water operators (DW), distribution system operators (DS), or wastewater operators (WW). Mail the completed page (or a photocopy) to: Florida Environmental Professionals Training, P.O. Box 33119, Palm Beach Gardens, Fla. 33420-3119. Enclose $15 for each set of questions you choose to answer (make checks payable to FWPCOA). You MUST be an FWPCOA member before you can submit your answers!

Earn CEUs by answering questions from previous Journal issues!

___________________________________ SUBSCRIBER NAME (please print)

Article 1 _________________________________ LICENSE NUMBER for Which CEUs Should Be Awarded

Article 2 _________________________________ LICENSE NUMBER for Which CEUs Should Be Awarded

If paying by credit card, fax to (561) 625-4858 providing the following information: ___________________________________ (Credit Card Number)

Contact FWPCOA at or at 561-840-0340. Articles from past issues can be viewed on the Journal website,

____________________________________ (Expiration Date)

Groundwater Nanofiltration Plant Addresses Color and Disinfection Byproducts for Flagler County

Dialing Down Disinfection Byproducts With Chlorine Dioxide Pre-Oxidation

Phillip J. Locke, Eric A. Smith, and Mark Ralph

Lance Littrell, Brian Gongre, Patrick Flynn, Domenic Gentilucci, Steve Romano, Rhea Dorris, and Gina Parra

(Article 1: CEU = 0.1 DW/DS)

1. Well No. 4 was used for pilot testing because it a. was the primary well in use at the time of pilot testing. b. is located next to a paved area. c. produced water of widely varying quality. d. produced water with the highest color and organics. 2. Pilot testing resulted in the highest percentage reduction in which of the following water quality parameters? a. Conductivity b. Iron c. Hardness d. Turbidity 3. Toward the end of pilot testing, second-stage iron fouling occurred, likely the result of a. overfeed of antiscalant b. inconsistent operation. c. booster pump failure d. biological foulants. 4. Final design of the two-stage nanofiltration (NF) treatment system will include ______________ to add pH and alkalinity to permeate. a. calcium hydroxide b. calcium carbonate c. carbon dioxide d. a partial bypass stream 5. In which of the following sections of the weighted treatment decision matrix did NF score higher than both lime softening and ion exchange? a. Costs b. Technical feasibility c. Operation and maintenance d. Schedule

(Article 2: CEU = 0.1 DW/DS)

1. Chlorine dioxide reactions in water do not form trihalomethanes or haloacetic acids because when chlorine dioxide oxidizes organic material it is reduced to ________, but does not chlorinate the resulting organics. a. chloramine b. ozone c. chlorite d. acetaminophen 2. A recent alternative method for generating chlorine dioxide is the reaction of _____________ with sodium hypochlorite and an acid. a. a strong base b. ferric chloride c. oxygen d. sodium chlorite 3. The U.S. EPA regulates chlorine dioxide as a primary disinfectant with a maximum residual disinfectant level of _____ milligrams per liter (mg/l). a. 0.6 b. 0.8 c. 1 d. 1.5 4. Initial testing revealed chlorine dioxide demand of the pilot test water to be closest to which of the following? a. 0.5 mg/l b. 0.8 mg/l c. 1.0 mg/l d. 1.2 mg/l 5. Chlorine dioxide has been shown to have _______________ oxidation potential and disinfection efficacy when compared to chlorine. a. equal b. somewhat less c. twice as much d. five times as much

Florida Water Resources Journal • August 2018



Giving Back Bill Young Chair, FSAWWA

s I’ve stated in previous articles, there are many positive reasons to be involved in professional organizations. With membership in FSAWWA you can increase your knowledge in our career field, expand your leadership capabilities, network with other professionals, and gain the advantages that go with all of these. What often goes unsaid is that participation in professional groups, like FSAWWA, also greatly benefits other people, and other communities around us. In my 25 years with the section, I have participated in many functions and events that directly benefit several groups. In this month’s article, I would like to look a little closer at the organizations and programs we give to.


Roy W. Likins Scholarship Fund Mr. Roy Likins was a lifelong member of AWWA and worked for the Palm Coast Utility Corporation for 16 years in many leadership roles. In addition to giving back to his community, he was also very involved in our Florida Section and served as section chair from 1980 to 1981. Mr. Likins died in 1991 at the very young age of 53. His service to our section, and to others less privileged, is recognized by naming the section’s scholarship program after this well-respected water leader. Mr. Likins would surely be proud of what we have done in his name. The Roy Likins Scholarship Fund began in 1988 and, to date, it has awarded 94 scholarships to undergraduate, graduate, and doctoral students, with total funding of over $350,000. Just last year, the fund gave over $40,000 in scholarships to


nine deserving students in the field of water. Moreover, we currently have 25 past Likins winners serving our section in some way. As a past awardee myself (in 1993), I can tell you how rewarding it is to see others give back to the section that has meant so much to our careers. The Roy Likins Scholarship Fund is a model for all other AWWA sections, and has been our section’s primary beneficiary for decades!

Water For People The purpose of the FSAWWA Water For People Committee is to facilitate the Florida water/wastewater industry’s participation in improving the lives of people around the world with respect to access to clean water and safe sanitation. The committee has represented the fundraising, volunteering, education, and outreach efforts of Water For People in the state of Florida since 1995. The committee fosters membership in FSAWWA by providing committee members in 12 Florida Section regions with opportunities for leadership development, participation in community fundraising activities, and cultural exchange through international impact tours. Through the efforts of the FSAWWA regions and board of governors, the section has contributed over $730,000 to Water For People since 1995. The organization has a chair or a point of contact in each of the 12 Florida Section regions where community outreach and fundraising activities are held every year, thanks to a strong base of volunteers in every region. A few of the past and future successful fundraising events in Florida include: S Region III – Wine For Water (10th Anniversary in 2018). Annual networking and fundraising event with over 300 attendees, raising over $50,000. S Region IV – Annual Water For People Fundraiser, July 19, 2018. Annual network-

August 2018 • Florida Water Resources Journal

ing and fundraising event. First year holding the event at Rock Brothers Brewing and expected to be very well-attended. Anticipate exceeding its fundraising goal of $5,000. S Region VI – Beer For Us. Annual networking and fundraising event showcasing local microbrewers in Palm Beach County. Raised over $6,000 during the last event in 2017. S Region VII – Annual Wine Tasting. Annual networking and fundraiser event. Attendance is over 200 people, raising over $8,000 per event, over the past three years. These fundraising events highlight the successful efforts that the volunteers work toward every year. It’s through the passion and teamwork in every region that the Florida Section committee was one of the top ten fundraising Water For People committees in fiscal year 2017 by raising $97,431. This year the committee is under the able leadership of Juan Aceituno.

The Water Equation Recently, you may have heard of AWWA’s Water Equation. This multifaceted program is the association’s “Investment in the Future of Water.” The Water Equation provides academic and operator scholarships, student memberships, and supports an amazing volunteer program, the Community Engineering Corps. Our section, with others, will partner with AWWA to invest in the future of water and our industry’s workforce through this exciting new program. I have no doubt that our section will stand out, as it always does! Thanks again for your support of the Florida Section. As you can see, we help many others and we should all be very proud of how we give back. You truly are changing lives! S

Florida Water Resources Journal â&#x20AC;˘ August 2018



Dialing Down Disinfection Byproducts With Chlorine Dioxide Pre-Oxidation Lance Littrell, Bryan Gongre, Patrick Flynn, Domenic Gentilucci, Steve Romano, Rhea Dorris, and Gina Parra Initial Considerations Today, the most commonly used disinfectants for potable water are chlorine and chloramine. The use of chlorine is increasingly subject to criticism due to its numerous reactions with organics and the respective regulations. Chlorine represents both safety- and health-related risks and effects and reacts quickly with organic matter to form disinfection byproducts (DBPs). Such effects can be mitigated by applying a disinfectant with different characteristics. As a potential alternative, chlorine dioxide (ClO2) is a strong and selective oxidizer and offers several advantages in treatment and distribution of drinking water. The ClO2 forms fewer halogenated DBPs and can be used at lower concentrations with shorter contact times to achieve equivalent disinfection than the concentrations and contact time required for chlorine and chloramine disinfection. It’s also less reactive to changes in pH than chlorine and has been proven more effective over a broader range of pH than free chlorine [1]. Since the 1970s, ClO2 has been implemented in distribution systems after the discovery of total trihalomethanes (TTHMs) and other DBPs that are still being discovered to date. It has been utilized in Europe and in the United States as both the primary disinfectant and pre-oxidant, with around 1,200 plants currently implementing its disinfection [1]. The selective reactivity enables ClO2 to control waterborne pathogens without reacting with organic DPB precursors. Unlike chlorine, ClO2 reactions in water do not result in the formation of TTHMs and haloacetic acids (HAA5) because “when ClO2 oxidizes organic material it’s reduced to chlorite, but does not chlorinate the resulting organics” [2]. It can be applied for a variety of water quality issues, including DBP formation control, taste and odor issues, or nitrification in the distribution system, especially in distribution systems where water age with long dead-end mains are a concern [2]. The use of ClO2 can be tailored to a specific facility’s need, and can be used for the primary disinfectant or as a preliminary oxidant, fol-


lowed by chlorine or chloramines. It has been shown to have five times stronger oxidation potential and disinfection efficacy than chlorine [3]. Realizing the impact of ClO2 on the regulatory challenges faced today, its applicability becomes very broad with potable water treatment and other means of disinfection. Recent studies have identified results indicating that ClO2 has significant potential to provide preliminary oxidation of organics prior to sodium hypochlorite disinfection, which has shown to reduce DBPs formed in the potable water distribution system. The ClO2 disinfection is an acceptable method of treatment within U.S. Environmental Protection Agency (EPA) regulations, as well as the Florida Administrative Code (F.A.C.), pursuant to the following: “All suppliers of water shall maintain a minimum free chlorine residual of 0.2 milligrams per liter, or a minimum combined chlorine residual of 0.6 milligrams per liter, or an equivalent ClO2 residual throughout their drinking water distribution system at all times.” Regulatory guidelines identify ClO2 as an acceptable method of inactivating viruses and bacteria to achieve 4-log virus inactivation and residual disinfection. The EPA regulates ClO2 as a primary disinfectant with a maximum residual disinfectant level (MRDL) of 0.8 mg/L. When dosed, ClO2 dissociates in water to form chlorite, which has a maximum contaminant level (MCL) of 1 mg/L. Controlling chlorite levels to comply with the MCL is one of the keys to successfully implementing ClO2. Chlorine Dioxide Generation Overview There are multiple ways to produce ClO2. Traditionally, it’s generated from the reaction of chlorine gas with sodium chlorite. Chlorine gas is not common to most municipalities due to the extensive safety risks associated with operation and storage. Accordingly, chlorine gasbased ClO2 generation is not applicable due to the aversion to the chlorine gas operational and safety concerns. Recently, alternative methods of generation have hit the market for ClO2 through the reaction of sodium chlorite with sodium hypochlorite and an acid, such as hydrochloric or sulfuric acid. The primary meth-

August 2018 • Florida Water Resources Journal

Lance Litrell, P.E., and Steve Romano, P.E., are project managers, and Rhea Dorris, E.I., and Gina Parra, E.I., are project analysts, with Kimley-Horn and Associates in Orlando. Patrick Flynn is vice president, Bryan Gongre is regional manager, and Domenic Gentilucci is area manager with Utilities Inc. of Florida in Altamonte Springs.

ods of ClO2 production are through a vacuum eduction generator, or through combining powder components to generate batch solutions, which contain stabilizers to minimize offgassing of ClO2 while stored. Regardless of the production method, ClO2 should be produced within a 0.1-0.5 percent solution, to reduce risk of an exothermic reaction. The ClO2 used in the pilot study was produced from vacuum eduction of three liquid components (sodium hypochlorite, hydrochloric acid, and sodium chlorite) through an onsite generator in a sidestream of water forming a 0.2-0.3 percent ClO2 solution. The equipment provided by this pilot testing application was supplied by Evoqua.

Pilot Study Pilot Overview Utilities Inc. of Florida (UIF) currently owns and operates the Lake Groves Water Treatment Plant (Lake Groves WTP) in the LUSI South service area in South Lake County. With the onset of the Stage 2 Disinfectant/Disinfectant Byproducts (D/DBP) Rule, UIF has made efforts to maintain compliance with DBPs through well blending, which places betterquality wells in service with high-organic wells to offset or minimize the DBP impact when the poorer quality wells are in service. This strategy lowered the TTHM levels, but the system has still periodically exceeded the regulatory limit of 80 parts per bil (ppb). The UIF currently utilizes sodium hypochlorite as the sole disinfectant for its storage and distribution system; as a result, the sodium hypochlorite reacts with the naturally occurring organics that produce TTHMs and HAA5. As such, UIF has sought al-

ternative methods of treatment, as well as disinfectants to achieve compliance with the Stage 2 D/DBP Rule. Utilizing the results from the laboratory testing to establish dosing parameters, UIF proceeded with 10 weeks of pilot testing at Lake Groves WTP utilizing ClO2 as a pre-oxidant within the water treatment process to achieve reduction of TTHMs formed in the distribution system. The results of this piloting effort confirm the overall reduction of TTHMs, as well as the system’s ability to maintain chlorite levels below the MCLs. During the full-scale pilot testing, laboratory data were collected for development of this report and subsequent verification for permit approval of the full-scale pilot testing implementation. The next step in the process was to demonstrate the laboratory effects on the full-scale utility system. A pilot testing approval package was completed and submitted to the Florida Department of Environmental Protection (FDEP). While the chemical has been used in the utility industry, a small number of utilities throughout the U.S. have used ClO2 for color, odor, and taste removal, with only a handful of utilities using it as a primary disinfectant. Accordingly, several questions and comments were discussed with FDEP prior to garnering the approval to proceed with the pilot. Following approval from FDEP, the full-scale pilot test was implemented at Lake Groves WTP. The overarching goals of the full-scale pilot study included a dosing of ClO2 and a gradual increase in concentration to determine the general range of effectiveness and vigorous field and laboratory testing of the treatment process, both during the pre-oxidant dosage and after, to ensure public safety, as well as compliance with the regulations. The utility and onsite staff completed routine efforts to operate, adjust dosages, and use the field work to obtain all the required samples. The staff ’s thorough analysis and consideration of the results proved very helpful in concluding the effect of each process adjustment. Upon approval from FDEP, the pilot was initiated at the Lake Groves facility for 10 weeks of testing and monitoring. The duration was selected to determine the optimum dosing rate to reliably maintain TTHMs and HAA5 below regulatory limits. The ClO2 was used in conjunction with sodium hypochlorite, which served as the primary and residual disinfectant within the utility distribution system. The sampling period allowed for a biweekly adjustment in dosing rate to determine the optimal dosage of ClO2 for each location. At the end of the pilot testing, the data were analyzed to confirm that the anticipated TTHM and HAA5 reduction in the dis-

tribution system was recognized, including the impacts of a varied dosage. Pilot Setup and Equipment The full-scale pilot injected ClO2 into the clearwell immediately following the forced draft aerators, and a parallel sodium hypochlorite injection was dosed within the same clearwell. The water was then pumped to the ground storage tanks (GST) where approximately 20 hours of storage resides under normal operational conditions. From the GST, finished water is pumped into the distribution system via the high-service pumps. Prior to the point of entry (POE), the water is continuously sampled for ClO2 residual and chlorite. The pilot program included the physical components to generate and inject ClO2 oxidant into the process stream. The physical equipment required to complete this pilot test includes the components as follows for the Lake Groves WTP site: S ClO2 Generation System – A ClO2 generation system rated for a maximum of 50 lb of ClO2 produced/hour was utilized for ClO2 production and injection. The generator system educts three chemical components into a potable water stream for safe and continuous ClO2 production. The generator was equipped with a control panel to adjust and monitor the dosage rate and was located in an enclosed area. The ClO2 is formed within the generator and was injected as a dilute solution. The ClO2 generation system was mounted on a stainless steel skid and consists of the following major components: • Water booster pump with downstream




pressure regulating valve and water rotameter to control the input water flow. • Three chemical feeds for sodium hypochlorite, sodium chlorite, and hydrochloric acid, each with rotameter for flow control. Chemical feed tubes were directly attached to chemical storage totes for vacuum suction. • Liquid jet venturi pump inductor (in situ). • Control panel to adjust and monitor dosage rate. ClO2 Chemical Precursors – 25 percent sodium chlorite, 12.5 percent sodium hypochlorite, and 15 percent hydrochloric acid were the three chemical precursors, which were vacuum-fed to the generation system to safely control the reaction and prevent unwanted byproduct formation. Sodium hypochlorite was already stored and used onsite. The sodium chlorite and hydrochloric acid were delivered in 265-gal chemical totes, which were attached to the generator feed by tubing inserted into the chemical tote. Sampling Stations – Several sampling taps located within the process stream were identified to pull grab samples of the treated water. Grab Sample Analyzer – One handheld analyzer for routine monitoring of ClO2 residual and chlorite (Palin-Test Analyzer was utilized for daily samples, as well as to confirm the online analyzer readings). Online Chlorite Sample Analyzer – One analyzer for continuous monitoring of chlorite levels at the POE to the distribution system. Continued on page 18

Figure 1. Chlorine Dioxide Pilot Test Schematic Florida Water Resources Journal • August 2018


Continued from page 17 S Online ClO2 Residual Sample Analyzer – One analyzer for continuous monitoring of ClO2 residual at POE. S Online ClO2 Monitoring and Control System – One control panel capable of receiving the analog signals from the online analyzers, pump controls, and operator interface with the control system The dosage of ClO2 was initiated at 0.8 parts per mil (ppm) dosing rate for the distribution system to achieve the desired TTHM and HAA5 reduction; it was then decreased to 0.6 ppm for two weeks and tested for the final four weeks at 1.0 ppm. Downstream of the injection point, both before and after storage, the ClO2 residual was monitored using a handheld ClO2 analyzer. Further monitoring in the distribution system included the POE, the average water age within the distribution system, and the extents

of the distribution system. The ClO2 generation system started and stopped in conjunction with Well #3, which has a 3,000-gal-per-minute (gpm) capacity and is the primary source of TTHM formation in the water from the Lake Groves WTP. Since the well pumping rate is fixed, the chemical dosing was paced on the constant flow rate and initiated directly with the well run times. The ClO2 levels were monitored a minimum of once per day within the eight hours of staffed operation of the treatment plant. The handheld probe identified the ClO2 levels that were used to confirm/regulate the feed rate of the ClO2. The online ClO2 residual analyzer, which sampled from the POE and was tied into the Lake Groves supervisory control and data acquisition (SCADA) system, had a predetermined maximum alarm set point of 0.6 ppm of ClO2 residual to ensure that the MRDL of 0.8 ppm was not exceeded. If the residual ppm level

Figure 2. Sample Location Map


August 2018 • Florida Water Resources Journal

reached 0.6 ppm of ClO2 residual, its generation system was programmed to turn off until manually reset by the operations staff. Due to the low initial dosage rate ClO2 residual at POE, it did not reach the 0.6 ppm maximum level. Figure 1 displays the pilot study process flow diagram for the Lake Groves WTP highlighting the chemical dosing and sample points within the treatment process. While the pilot operations were ongoing, it was imperative to monitor the performance, as well as any concerning parameters within the treatment plant and throughout the distribution system. A monitoring and sampling plan was implemented throughout to study to ensure that the performance could be quantified and public health protected. Distribution System Sampling Distribution system samples were conducted weekly throughout the pilot study and were analyzed in the Orlando Utilities Commission (OUC) Water Quality Laboratory for TTHM and HAA5 concentrations. The distribution system DBP samples are imperative to the pilot study because they measured the ability of ClO2 to delay/eliminate DBP formation. The distribution system locations were selected to provide a DBP formation curve demonstrating the beginning to the extents of the distribution system. The average water age of each location was used to compare the distribution system results to the baseline chlorinated TTHM formation curve. Water age was determined from performing an analysis in the distribution system hydraulic model during existing average daily demand (ADD) conditions. The following sampling locations were utilized for the TTHM and HAA5 distribution system analysis: S POE – Represents the point that the disinfected and treated water enters the distribution system. (Approximate water age = 1 day/24 hours) S Residual Site 1 (R1) – Represents the average residence time location, and was taken off a potable water sample tap at the entrance to the Savannas neighborhood, north of the Lake Groves WTP. (Approximate water age = 1.7 days/40 hours) S Residual Site 2 (R2) – Represents the maximum residence time location, and is one of the FDEP Stage 2 D/DBP Rule compliance locations. (Approximate water age = 2.25 days/54 hours) S Lake Louisa WTP (Connected Consecutive Water System) – Represents the point where the water from Lake Groves WTP enters the LUSI North service area by feeding into the Continued on page 20

FWPCOA TRAINING CALENDAR SCHEDULE YOUR CLASS TODAY! August 13-17 ....Fall State Short School ..........................Ft Pierce

September 10-13 ....Backflow Tester ......................................Osteen ............$375/405 28 ....Backflow Tester Recerts*** ..................Osteen ............$85/115

October 8-10 ....Backflow Repair ....................................Osteen ............$275/305 15-18 ....Backflow Tester* ....................................St Petersburg ..$375/405 15-19 ....Wastewater Collection C, B..................Orlando ..........$225/255 26 ....Backflow Tester Recerts*** ..................Osteen ............$85/115 26 ....Backflow Tester Recerts........................Pensacola........$85/115

November 5-8 ....Backflow Tester ......................................Osteen ............$375/405 12-14 ....Backflow Repair* ..................................St Petersburg ..$275/305 12-16 ....Reclaim Water Feld Site Inspector ......Osteen ............$350/380 16 ....Backflow Tester Recerts*** ..................Osteen ............$85/115 Course registration forms are available at For additional information on these courses or other training programs offered by the FWPCOA, please contact the FW&PCOA Training Office at (321) 383-9690 or * Backflow recertification is also available the last day of Backflow Tester or Backflow Repair Classes with the exception of Deltona ** Evening classes

You are required to have your own calculator at state short schools and most other courses.

*** any retest given also Florida Water Resources Journal â&#x20AC;¢ August 2018


Continued from page 18 Lake Louisa GST. (Approximate water age = 2 days/48 hours) The distribution system analysis sample locations are displayed in Figure 2.

Results and Observations

Figure 3. Well #3 Raw Total Trihalomethane Formation Potential

Figure 4. Chlorine Dioxide Dosage Versus Demand Curve

Baseline Formation Potential The University of Central Florida (UCF) Environmental Systems Engineering Institute (ESEI) conducted an evaluation in April 2016 of the byproduct formation for all the wells supplying the LUSI North and South service area. From the testing, a sample was pulled from Well #3, which identified it as the biggest contributor to the facilityâ&#x20AC;&#x2122;s water quality challenges; specifically, Well #3 testing revealed high TTHM formation potential, as shown in Figure 3. The Well #3 TTHM concentrations reached 130 ppb at a disinfection contact time of 96 hours (~4 days), which is expected to be above the Stage 2 D/DBP Rule limit set when blended with the other wells supplying this facility. The 80-ppb MCL is exceeded at a low water age of approximately one day. The ESEI also reported an HAA5 concentration of 50.95 ppb at a disinfection contact time of 96 hours. The Well #3 formation curve serves as a baseline for comparison with the delayed formation when utilizing ClO2 pre-oxidation within the pilot testing. Chlorine Dioxide Demand Testing Preliminary onsite testing was completed to determine the ClO2 demand on the aerated raw water on June 21, 2017. Since ClO2 was planned to be injected downstream of the packed tower aerators during full-scale pilot testing, the preliminary sample was aerated in the laboratory before demand testing was conducted. The sample was divided and dosed with five different concentrations of ClO2, ranging from 0.5 to 1.5 ppm. The samples were stored in a dark container to prevent ultraviolet degradation for 45 minutes. The ClO2 demand (calculated from subtracting the ClO2 residual after 45 minutes from the initial dosage rate) leveled off at approximately 0.86 ppm at an initial dosage of 1.2 ppm. The demand versus dosage curve that resulted from the ClO2 demand testing is shown in Figure 4. This set the standard for the pilot testing to begin at an initial dosage of about 0.8 ppm.

Figure 5. Total Trihalomethane Results


August 2018 â&#x20AC;˘ Florida Water Resources Journal

Disinfection Byproduct Reduction The full-scale pilot TTHM concentrations are compared to the chlorinated formation potential (containing no ClO2) in Figure 5. At a

time of approximately 24 hours, the POE ranged from 24-40 ppb, as compared to the formation curve that did not contain ClO2, which had already exceeded the 80 ppb limit. The 40hour time was represented in the distribution system by site R1, and ranged between 27-50 ppb. The results at site R2 ranged from 37â&#x20AC;&#x201C;52 ppb, a 50 percent reduction from the baseline formation curve. These data confirm the effectiveness of using ClO2 as a preliminary oxidant to deter the formation of TTHMs. At the minimum 0.6 ppm dosage tested during the pilot, the TTHM concentrations were approximately 10 percent lower than the maximum dosage of 1.0 ppm, showing a small difference in formation for the minimum and maximum dosing rates. The HAA5 distribution system results, displayed in Figure 6, showed a range of 12-22 ppb at the POE, 14-22 ppb at the average residence time location, and 21-30 ppb at the maximum residence time location. The HAA5 remained at least 50 percent below the 60 ppb limit through the duration of the pilot. The HAA5 concentrations were approximately 40 percent lower than the baseline comparison value of 50.95 ppb reported by ESEI. Regulatory Compliance Sampling Throughout the pilot study, the ClO2 residual at the POE remained at 0.06 mg/L and less, as shown in Figure 7, which is near zero and significantly below the MRDL of 0.8 mg/L. Many ClO2 residual readings were recorded at 0.01 mg/L, which is the lower detection limit of the online analyzer. These results confirm the hypothesis that a majority of the ClO2 was consumed prior to reaching the entrance of the distribution system. Minimal ClO2 residual is expected with pre-oxidation due to the small initial dosage needed and the relatively high organics found in the raw water. The chlorite levels, displayed in Figure 8, were maintained at less than 0.25 mg/L, well below the 1.0 mg/L MCL. Since chlorite is formed from the aqueous dissolution of ClO2, the chlorite concentration increases as ClO2 demand is consumed. The POE samples tested were the maximum amount of chlorite recorded within the testing analysis.

Figure 6. Haloacetic Acids Results

Figure 7. Chlorine Dioxide Residual at Point of Entry

Conclusions Based on the full-scale pilot study, the following improvements are recommended: S Install pre-oxidation ClO2 generation system to reduce TTHM and HAA5 concentrations in conjunction with the current disinfectant used. Continued on page 22

Figure 8. Chlorite at Point of Entry Florida Water Resources Journal â&#x20AC;˘ August 2018


Continued from page 21 S Initiate the ClO2 generation system at a design dosage of 0.8 ppm (26.67 lb/day). S The generator for permanent installation can be wall-mounted in the existing sodium hypochlorite building, with the hydrochloric acid and sodium chlorite bulk storage tanks stored outside of the building under a covered structure. The following timeline is recommended for full-scale installation: S May – June 2018: Construction permit application and approval S June – December 2018: Construction S January 2019 – Stage 2 D/DBP Rule compliance with all production wells The estimated capital cost for construction of a permanent system is approximately $200,000. The yearly operating costs include three chemical generation components: sodium hypochlorite, hydrochloric acid, and sodium

chlorite. The estimated annual cost of a permanent ClO2 generation system is approximately $66,000 per year if the system were to run constantly at the 4.32-mil-gal-per-day (mgd) or 3,000-gpm well capacity, which equates to an annual operating cost of $15,600 per mgd, or $0.0428/1,000 gal of water produced. Table 1 compares the approximate cost of a ClO2 system to three other DBP precursor removal methods: granular activated carbon (GAC), ion exchange, and reverse osmosis (RO) membranes. The costs for the three additional options are consistent with the information presented in the 2016 “Lake Groves Disinfection Reduction Report.” An average daily flow (ADF) of 3 mgd based on planned demands for the service area was used for annual cost projections; the capital cost differential for ClO2 versus the treatment identified is significantly less. In addition, ClO2 operating costs are also approximately 25 percent of the cost for RO membranes, which is the next cost-effective option. The cost projections over a 20-year period

Table 1. Chlorine Dioxide System Cost Comparison

Figure 9. Capital and Operational Cost Comparison


August 2018 • Florida Water Resources Journal

are shown in Figure 9. The long-term cost of ClO2 is $12 million less than the next lowest option; furthermore, the system only requires Well #3 for operation, and Wells #1 and #2 can return to backup operation. Recommendations The ClO2 proved to be highly effective at minimizing DBP formation, while saving capital costs compared to other treatment upgrades; however, ClO2 is sparsely used for potable water applications, so it’s imperative to fully understand the process before investigating its use. The following recommendations are based on lessons learned from the pilot study and extensive efforts of ClO2 testing at other facilities prior to this full-scale study. S The ClO2 is proven to be an effective tool to maintain compliance with the Stage 2 D/DBP Rule, but it’s still recommended to perform field and laboratory testing to verify the compatibility with the water characteristics. A fullor pilot-scale study is recommended prior to installation of a permanent ClO2 system to evaluate the effects within the distribution system. The goals of the pilot study would be to reveal the effects of ClO2 on a system’s specific water quality and identify optimal ClO2 dosing for maximum cost savings. S It’s important to gain understanding and consensus from state and local regulators and to remain in compliance with all water quality regulations while performing a ClO2 pilot study. Chlorite levels in the distribution must be monitored regularly and maintained below the MCL. It’s recommended to ensure that the ClO2 and chlorite samples are being accurately assessed from either a laboratory or a handheld sample analyzer. Inaccurate test results and wrongly reported concentrations can affect regulatory compliance and cause unnecessary public concern. S It’s recommended that the available options for ClO2 generation be reviewed and understood. Several factors are important when understanding generation options, including operator training and availability, goal usage of ClO2, redundancy needs, and chemical safety. Moreover, the aspects of each generation system need to be compatible with the process application and utility production conditions. Generators often produce ClO2 on demand; however, ClO2 storage is not often recommended for these in situ generation units. S Proactive and direct public communication is recommended before ClO2 is used in treatment processes. If utilizing ClO2 as a disinfectant, proper notification is required, similar to the transition from chloramine to

chlorine disinfection. Although ClO2 technology is not “new,” the public may be concerned hearing about the use of an unfamiliar chemical. It’s important to emphasize the benefits of ClO2 and compare its safety to typical disinfectants. Final Considerations All in all, careful consideration should be given to the implementation of ClO2 within a water production or distribution facility. While the chemical is effective in maintaining disinfectant residuals, as well as improving aesthetics in distribution system water quality, the appropriate process addition may be as a pre-oxidant, rather than as the primary disinfectant. The use of ClO2 has shown promise as a strong disinfectant chemical for other utilities aspiring to reduce DBPs without incurring significant capital cost associated with high-end treatment or the routine distribution system maintenance challenges associated with chloramines. As a viable pre-oxidant or alternative disinfectant, it should be considered when these DBP or distribution system challenges are present.

Acknowledgments Special thanks to those who made this project possible: the Utilities Inc. of Florida management team and its operations staff members who supported the application and provided extensive hands-on testing analysis for the data collection and continuous analysis of the field data; Evoqua, which provided the chlorine dioxide generation equipment, operations staff training, and generator oversight; and the FDEP staff members who participated in prepilot testing review meetings, as well as reviewed and approved the request to conduct the pilot study.

References [1]




Gates, Don, et al., 2011. “State of the Science of Chlorine Dioxide in Drinking Water.” American Water Research Foundation. Holden, Glenn W., 2017. “Chlorine Dioxide Preoxidation for DBP Reduction.” Journal - American Water Works Association, vol. 109, pp. 36–43. doi:10.5942/jawwa.2017.109.0089. Masschelein, W. J., 1979. Chlorine Dioxide. Ann Arbor Science, Mich. James, Cheryl, et al., 2004. “‘Relationships Between Oxidation-Reduction Potential, Oxidant, and PH in Drinking Water." S

Low Concentrations of Silver Can Foil Wastewater Treatment Research at Oregon State University (OSU) has shed new light on how silver nanoparticles, an increasingly common consumer product component, can potentially interfere with the treatment of wastewater. The findings suggest conventional toxicity testing methods for silver concentrations at treatment plants may produce results that yield a false sense of security. The research is important because if silver, which has broadspectrum antibacterial properties, thwarts the work of a plant’s beneficial bacteria, then too many nutrients end up in waterways. This in turn can lead to eutrophication, which is an overabundance of nutrients in a body of water that results in an explosion of vegetation, such as an algae bloom, and a squeezing out of animal life due to a lack of oxygen. "Silver nanoparticles are being incorporated into a range of products, including wound dressings, clothing, water filters, toothpaste, and even children's toys," said corresponding author Tyler Radniecki, an environmental engineering assistant professor at OSU. "The nanoparticles can end up in wastewater streams through washing or just regular use of the product." The work by Radniecki and collaborators in the college of engineering at OSU looked at silver nanoparticles, the ionic silver they release, and an ammonia-oxidizing bacterium, Nitrosomonas europaea. Ammonia-oxidizing bacteria (AOB) are crucial because they convert ammonia to nitrite to begin the process of getting one of those nutrients (nitrogen) out of the wastewater. The study looked at both free-floating, or planktonic, N. europaea, and also the biofilms they create. The OSU research confirmed earlier observations that biofilms are better able than planktonic bacteria to ward off silver's effects. "Biofilms showed higher resistance for

multiple factors," said Radniecki. "One was simply more mass of cells, and the top layer of cells acted like a sacrificial shield that allowed the bacteria below not to be inhibited. Slow growth rates were also a protection from silver toxicity because the enzymes that silver prevents from turning over aren't turning over as frequently." More importantly, the work unveiled a new wrinkle: the inhibition of AOB's ammoniaconversion ability is more a function of silver exposure time than the level of silver concentration. "Most of the studies investigating the inhibition of wastewater biofilms by nanoparticles have been conducted in shortterm exposure scenarios, usually less than 12 hours," Radniecki said. "Also, they've used an equal amount of time for hydraulic residence and sludge retention." The problem with that, he explains, is that in a treatment plant that uses biofilms, the sludge retention time—the time the bacteria are in the plant—will be much greater than the hydraulic residence time; i.e., the time the wastewater is in the plant. "That allows, over time, for the accumulation and concentration of metal contaminants, including ionic silver and silver nanoparticles," said Radniecki, whose worked involved exposure times of 48 hours. "The immobilized biofilm cells are exposed to a much greater volume of water and mass of contaminants than the planktonic cell systems. This means that the results of short-term exposure studies may fail to incorporate the expected accumulation of silver within the biofilm; wastewater plant monitors might then be underestimating the potential toxicity of long-term, low-concentration exposure situations." The National Science Foundation supported this research, and findings were published in Chemosphere. S

Florida Water Resources Journal • August 2018


Florida Water Resources Journal â&#x20AC;˘ August 2018



Your Region Needs You! Mike Darrow

ism in the discipline that you work in, whether you’re just starting out on your career path or are a seasoned veteran.

President, FWPCOA

s you know, the last two letters of FWPCOA stand for “operators association.” We’re a unique organization divided into 13 regions around the wonderful state of Florida. I’ve been a member of Region X for a lot of years now. I enjoy hearing about what’s going on in our industry and how the association can make a better future for our members, and it’s gratifying to know that FWPCOA is again growing in membership. Recently, I attended meetings in both Region VII and Region X. They were great meetings and I learned quite a few things at each of them, but one thing is clear to me: your region needs your ideas, skills, help, and professional-



Get Involved When most regions meet, the attendees discuss current issues and new technologies relevant to our trade to help inform and train operators, technicians, and coordinators as they advance in their careers. At the regional level, industry professionals and vendors get together to share ideas and network to help advance their own education and help their workplaces be more efficient, professional, and effective. We need to be better at passing the torch on what we have learned in our industry to new members. This is something that you can participate in and attending a local meeting is a good start. At the meeting you can share your thoughts on: S What types and disciplines should be taught locally? You can then participate to find re-

August 2018 • Florida Water Resources Journal

sources to put the classes on. This will help strengthen our skills and the tools of our trade. S What kind of continuing education unit (CEU) courses would be helpful to the regional membership? Some regions have opening for officers and trustees right now, so you can plug in, participate, and determine the courses needed. Some of our goals at FWPCOA are to protect the public health and the environment of our great state; to work on behalf of operators of all types, like technicians, coordinators, and those in other disciplines, by training them and making their work tasks more professional; and representing operators as a whole to the state. I ask you to please bring these types of ideas to the meetings to share. Another observation I have is that regions are in need of trainers or instructors to help teach at classes locally or at short schools, and one day, at CEU classes. Your profession is in

need of teachers for voluntary certification classes like industrial pretreatment, stormwater, wastewater collection, water distribution, water reclamation distribution, and customer service. If you are interested, it could really make a different to a person getting started in our profession. If interested, please fill out the instructor biographical form available on our website at and send it to the training office at to get started. Each member has a login for our website; for help with your password contact our webmaster, Walt Smyser, at The regional level is also a great place where you can socialize. Maybe this is what you like to do—so get involved here, too. The regions set up social and fun events like fishing tournaments, golf outings, family picnics, baseball games, and Christmas parties. Most of the training put on throughout the year by the regions helps pay for the social events in that region. We have a lot in common, so why not get together to share your thoughts on our profession, and also have some fun!

Indirect and Direct Potable Reuse Survey on FWPCOA Website Future water supply needs and new technology for water and wastewater are merging in a new area for our industry: direct potable reuse. This is helping to close the gap on the “One Water” concept, focusing on the recycling and reuse of this precious resource. Operators will have a key role in this task—as they do now—in operating water and wastewater facilities. The scarcity of water, its varying availability, and the cost to treat potable water is driving this issue. In the near future, as development continues in our state and elsewhere, the demand for water will far exceed supply. This is why potable and indirect potable water reuse is a good choice. When the good, less-expensiveto-treat water is gone, or allocated to other sources, options for supply will be limited. Why not recycle water to high quality and reuse it for years to come? As we know, water is extremely hard to create—two combustible gases mixing to form H2O! This means that the water we have today is water from eons and eons ago. Water has three forms: liquid, gas, and solid. The liquid kind— drinkable water—is becoming harder to find in some regions of the world. Most of our water, of course, is saltwater, and much of the water around the world is polluted, so it’s our job every day is to remove the contaminants and make it safe to use again and again. The future of water reuse is here. Rules and guidelines are now being determined by a state

commission to figure out a path for the industry to follow. At some point, operators and technicians should get involved in this process, so I encourage you to get involved with your local utility, or at any level in the state regulatory process, and get in on the conversation. I had the pleasure of making a presentation at the Potable Reuse Commission in June, providing information about how the operators, mechanics, technicians, and coordinators in the industry will work in our great state to help furnish future supplies of water. Many of us have

been involved for years in water reclamation, and reuse is just the next level of advanced treatment. We need your ideas and thoughts on the subject. Please participate in the survey on our website and provide the important input that will affect your own future, and the future of the industry. Thank you for your involvement in our regions and your input on the survey—and remember to go with the flow! S

Bonita Springs Utilities Announces Water Conservation Poster Contest Student Winners Bonita Springs Utilities Inc. (BSU) has announced its winners of the Drop Savers water conservation poster contest: S Genevieve Harper, Bonita Springs Elementary first-grade student, Division 1 S Edwin Nicolas, Spring Creek Elementary third-grade student, Division 2 S Emely Meneses, Spring Creek Elementary fifth-grade student, Division 3 Conducted in partnership with the Florida Section of the American Water Works Association, BSU coordinated with Bonita Springs and Estero elementary schools to encourage students to create a poster depicting a water conservation idea in slogan form, drawing form, or both. The contest helps students learn about and promote water awareness and the importance of water conservation in their daily routines. The winners were chosen based on the water conservation message,

creativity, and originality depicted through their artwork. “It benefits all of us when our youth learn early about the importance of water conservation,” said John R. Jenkins, BSU executive director. “We are proud to support the efforts of our local parents and teachers to educate students about responsible water use.” Each of the BSU winners received award certificates and $25 gift cards to both Prado Stadium 12 and Royal Scoop. The winners also received a Drop Savers 201819 school calendar that features their artwork, water conservation kits, and certificates of participation. In addition to being BSU’s Division 3 winner, Emely Meneses won second place for Division 3 at the FSAWWA state level. She also received a $75 Amazon gift card and note cards printed with her winning artwork. S

Pictured left to right are Genevieve Harper, Emely Meneses, and Edwin Nicolas.

Florida Water Resources Journal • August 2018



Leaders are Learners Kristiana S. Dragash, P.E. President, FWEA It‘s August, and that means back to school! But back to school isn’t just for kids. In our profession, continuing education is also a necessity. And for you professional engineers, don’t forget: February 2019 is license renewal time, which means you’ll need 18 continuing education hours to qualify. Luckily, FWEA has you covered with some great seminars coming up in the fall of 2018 and winter of 2019. The Utility Management Committee (UMC) and Florida Benchmarking Consortium (FBC) will be holding a seminar, “Best Practices in Benchmarking,” from 8:30 a.m. to 4 p.m. on Nov. 2, 2018, at the South Cross Bayou Water Reclamation Facility Training Center in Pinellas County. Thank you to Pinellas County for welcoming FWEA into your facility, and to all of the utilities that work with us in providing professional development opportunities for our members and the industry.


Also coming up this fall is a Wastewater Process Committee (WWPC) seminar in southeast Florida. Last, but certainly not least, be on the lookout this fall for the first seminar brought to you by the newly formed Contractors Committee and the Manufacturers and Representatives Committee (MARC). In early 2019 there will also be a few seminars you won’t want to miss, including those put on by the Water Resources, Reuse, and Resiliency (WR3); Biosolids; and Air Quality committees. Though this will be too late to use for your 2019 renewal, the Water Environment Federation (WEF) will be coming back to Fort Lauderdale for its Biosolids Conference in May 2019, and the Florida Water Resources Conference (FWRC) is always a wonderful way to network with colleagues and get your continuing education requirements taken care of (April 14–17, 2019, in Tampa). Hopefully you get a chance to attend the presentations and don’t spend your time running from one end of the exhibit hall and conference center to the other, like myself. But hey, as an officer of FWEA, per 61G15-22.003 F.A.C., I can claim a couple of hours toward my continuing education requirement. And guess what? So can you, as long as you’re ei-

August 2018 • Florida Water Resources Journal

ther an officer of the organization or you actively participate on a committee in the organization. Yet another reason to get engaged in one of the many committees I listed! You can look at these requirements as tasks to check off of your to-do list, or you can chose to see them as ways to improve yourself and become an even better professional. In this age of ever-evolving technology, there is always something new to learn. These “requirements” are actually there to help us, but sometimes it takes something mandatory for us to take a step back from our overflowing schedules and deadlines and learn something new! I encourage you to pick a couple of the committees I mentioned and attend one of the seminars that resonates with you; or even better, join the committee and help to plan the seminar. A common theme among the self-improvement experts I follow on social media and listen to on audiobooks is that you cannot reach your next level of success until you are ready for it. In other words, you have to become that person who can handle that next level of success before it becomes available to you. So, next time you feel stuck, instead of looking for external causes and solutions, look within yourself and determine what ways you need to improve to ascend to your next level and reach your goals. S


Can Machine Learning and Geostatistics Overcome Lack of Data in Assessing Recovery of Water Levels and Ecological Conditions at Unmonitored Wetlands and Lakes? Dan Schmutz, Stephanie K. Garvis, Danny Goodding, and Christopher Shea ampa Bay Water is the largest wholesale supplier of water in the state of Florida, serving more than 2.5 million people in the Tampa Bay Area (Figure 1). Some of the wellfields in Tampa Bay Water’s system have been pumping groundwater for over 50 years. Groundwater continues to be a vital part of the Tampa Bay region’s water supply, with more than 50 percent of the regional supply coming from wellfields (Tampa Bay Water, 2017). Though groundwater pumping has been drastically cut back, previous pumping from these wellfields contributed to lower water levels in some of the region’s lakes and wetlands, which led to deleterious ecological changes.


In 1998, Tampa Bay Water gained ownership and control of all of the regional wellfields in the Tampa Bay area. The Southwest Florida Water Management District (SWFWMD) issued a new permit to Tampa Bay Water that consolidated the permits for 11 of these wellfields located in Pasco, northern Hillsborough, and northeast Pinellas counties. This new permit, known as the “consolidated permit,” lowered the permitted annual average pumping limit for these 11 wellfields from 192 mil gal per day (mgd) to 90 mgd. Tampa Bay Water currently operates these wellfields as an interconnected system at this lower pumping limit to promote environmental recovery near the wellfields.

Figure 1. Tampa Bay Water existing groundwater and newer surface water supplies.


August 2018 • Florida Water Resources Journal

Dan Schmutz, M.S., is chief environmental scientist, and Stephanie K. Garvis, M.S., and Danny Goodding, M.S., are environmental scientists with Greenman-Pedersen Inc. in Orlando. Christopher Shea, M.S., P.W.S., is a senior environmental analyst with Tampa Bay Water.

Background Tampa Bay Water is required by Special Condition 11 of Water Use Permit (WUP) No. 20011771.001 to evaluate “the recovery of water resource and environmental systems attributable

Figure 2. Unmonitored sites of concern and Tampa Bay Water wellfields.

to reduction of…withdrawals…to 90 mgd.” As described in the permit recovery assessment work plan and schedule (Tampa Bay Water, 2012), a key issue to be resolved is how “wetland health (or recovery) criteria may be applied to wetlands lacking hydrologic data” (i.e., “unmonitored” wetlands). Tampa Bay Water (2013) previously defined areas of investigation for the recovery analysis using hydrologic modeling output and geographic information systems (GIS) analyses to produce a composite 2-ft surficial aquifer system (SAS) drawdown (DDN), representing the maximum of historical pumpage and several possible future scenarios, both scaled to 90 mgd. A recovery assessment GIS deliverable previously prepared by Greenman-Pedersen Inc. (GPI), and provided to Tampa Bay Water in January 2016, found 684 unmonitored sites occurring within the 2-ft SAS DDN contour, consisting of 675 wetlands and nine lakes, shown in Figure 2. (Subsequent analyses, based on revised modeling output available after the completion of this study, have increased the number of unmonitored sites to 749, and results for the entire set will be reported in a future publication.) Tampa Bay Water is being assisted by GPI in developing proposed methods for estimating ecological and hydrological conditions at unmonitored sites. This article describes these proposed recovery assessment methods, as well as preliminary results of analyses performed to facilitate method development. Following approval of the proposed methods, it’s anticipated that they will be applied in a future project phase to aid in the assignment of unmonitored sites to appropriate recovery assessment (RA) status bins. These bins document wetland/lake conditions relative to approved recovery metrics, as well as evidence of trends towards recovery, and they consist of the following categories: S Never impacted S No cutback, meets metric S Recovered S Improved, not fully recovered S Not fully recovered, continuing wellfield impact S Impacted due to other causes S More detailed analysis needed Conceptually, the problem of assessing unmonitored wetlands is one of statistical interpolation. Specifically, there is a need to develop defensible approaches for transferring information from nearby sites with known recovery assessment statuses to unmonitored sites. The term “nearby” might imply physical proximity, but it could also imply proximity in multivariate space (i.e., statistical “nearest neighbors”). Determination of appropriate spatial support

Figure 3. Conceptual model of regression-kriging approach to predicting data at unmonitored sites.

(e.g., “Over what distance is recovery status correlated?”) is a subset of the overall problem of determining how recovery varies among sites that are close in a statistical sense. Development of statistical models to allow inference at unmonitored wetlands requires the development of adequate datasets from nearby monitored wetlands collected from an appropriate time period (e.g., the period after wellfield production was reduced to 90 mgd, or postcutback). In seeking to develop these methods, it’s anticipated that the methods might vary by wetland community type (e.g., isolated cypress) or surrounding soil classification (i.e., xeric or mesic) based on previous findings that water levels in wetlands in different soil settings behave differently (GPI, 2016). Methods also were anticipated to be hierarchical in nature, meaning that broad screening tools would be proposed as a first cut to classify, or bin, wetlands using the least amount of new data collection. The unmonitored wetlands and lakes of concern occur primarily in eight regions within the northern Tampa Bay (NTB) area associated with various wellfields (Figure 2): S Eldridge-Wilde (ELW) S Northwest Hillsborough (NWH) S Section 21 (S21) S Morris Bridge (MBR) S Cypress Bridge (CYB) S Cypress Creek (CYC) S Cross Bar Ranch (CBR) S CYC/CBR interwellfield area More than 400 wetlands, lakes, and connected systems with water-level data (i.e., monitored sites) also occur in and around these areas, providing a potential basis for statistical interpolation to the unmonitored sites. Some ecological data are available for selected unmonitored sites, including wetland health assessment (WHA) data from previous SWFWMD studies. This present study also investigates the utility of

several regional datasets believed to have some potential for supporting predictions of wetland water levels at unmonitored sites, including SAS DDN, Upper Floridan aquifer (UFA) DDN, and surrounding soil classification. Given the large number of potential variables that might be useful for interpolating water levels at unmonitored sites, the discipline of machine learning, a field of computer science that develops algorithms that can learn from and make predictions on data (Alpaydin, 2009), was used to help develop robust statistical models, which are those that are expected to perform well on new data, meaning they are not “overfit” to unique aspects of a particular dataset. The danger in overfitting is that a model that appears to perform well on past data (or data located in certain spatial areas) will perform poorly in the future (or in other unstudied spatial locations) because the analyst mistakenly fit the model to noise rather than signal in the development dataset (James et al., 2017). In the past three decades, there has been increasing recognition of the value of hybrid spatial interpolation approaches to predicting values at unmonitored locations. These hybrid techniques combine two conceptually different but complementary techniques: analysis by multiple linear regression (or other machine learning algorithm) to predict based on aspatial auxiliary variables, and spatial interpolations based solely on values of the known points and their spatial autocorrelation characteristics (e.g., kriging). Typically, these hybrid techniques provide more accurate predictions than either single approach (Hengl et al., 2007). Hybrid spatial interpolation techniques have been applied successfully to problems such as water-table mapping (Desbarats et al., 2002), interpolation of soil properties (Odeh et al., 1994; Hengl et al., 2015), and estimation of rainfall (PardoIguzquiza, 1998), among many others. Continued on page 38

Florida Water Resources Journal • August 2018


Continued from page 37 Regression-kriging (RK) is a general term used to describe a hybrid spatial interpolation technique that may involve the separate fitting of an aspatial model (e.g., multiple linear regression, random forest, etc.) and subsequent kriging of residuals from the model (Hengl et al., 2003). Key assumptions of the technique are that residuals are normally distributed with constant variance and the values of the auxiliary variables are known at all locations needed for prediction.

Methods In this study, RK was used for development of the best model to predict historical median (2008-2014) wetland water levels at unmonitored sites by using a combination of aspatial and spatial information from nearby monitored sites. The technique required development of the best possible multiple linear regression model (i.e., selecting most predictive variables while avoiding overfitting), and then examining the residuals from that model for spatial autocorrelation. If positive spatial autocorrelation was found to be present, then kriging of the residuals would be expected to result in improved predictions for the unmonitored sites (Hengl, 2009). In other words, if there is a tendency for the initial aspatial model to predict median wetland water levels too high or too low in spatial clusters, it’s implied that unmeasured but spatially autocorrelated factors—not included in the ini-

tial model—are affecting the outcome. Although those factors aren’t known, their effect can be modeled at unmonitored sites using the technique of kriging, which is a geostatistical procedure that predicts how similarity in the residuals changes with distance. Kriging estimates deviations at unmonitored sites by a weighted averaging of nearby residuals. With positive spatial autocorrelation, residuals near each other will tend to be more similar than residuals farther apart. In summary, RK involves spatially interpolating residuals from an aspatial model using kriging and adding the results to the predictions from the aspatial model. Conceptually, the aspatial regression predictions for unmonitored sites will be adjusted up or down based on the residual deviations of nearby sites (Figure 3). The focus of hydrologic prediction at the unmonitored sites was the median water level for years 2008-2014 relative to a high water mark at each site known as the historical normal pool (HNP). The offset of the median water level relative to the HNP—known as the normal pool offset (NPO)—was selected to represent the most appropriate surface water hydrologic correlate of wetland health, based on the results of scientific investigations performed in the NTB area, including the development of minimum levels for isolated cypress-dominated systems (SWFWMD, 1999). The NPOs represent the “offset” of wetland water levels from a reference elevation of historic inundation (i.e., the normal

Figure 4. Hydrograph documenting increasing six-year median water levels following production cutbacks.


August 2018 • Florida Water Resources Journal

pool). The HNP offsets can be related to historic conditions and they are not affected by differing wetland depths, unlike hydroperiods. Also, the use of a local wetland reference datum, like HNP, allows for the analysis of a large number of sites occurring across a wide range of absolute elevations on a common scale (i.e., ft below HNP). The seven-year time period of 2008-2014 was considered appropriate because it represented a postcutback and operationally stable pumpage configuration, with a range of rainfall conditions, although this time period may be conservative (i.e., underpredicting long-term median wetland water levels), depending on the time frames involved in the reduction of groundwater pumpage on the local scale, and the subsequent aquifer response. The interpolation methods involved: S Identifying all sites (e.g., lakes, isolated wetlands, and connected wetlands) with HNP elevations and water-level data for the period of 2008-2014. S Excluding or truncating erroneous data. S Determining appropriate groups for interpolation. S Determining auxiliary variables potentially useful for aspatial prediction of NPOs. S Developing the best aspatial multiple linear regression model using an information criterion-based search through all possible models. S Performing RK to fit a variogram model to explain spatial-autocorrelation in residuals from the aspatial multiple linear regression model. A small number of sites were excluded due to groundwater augmentation or data problems. There were 309 monitored sites (wetlands and lakes) with adequate data to calculate a median offset for 2008-2014 for use in developing a model to predict water levels at the unmonitored sites. Wetland and lake water levels for monitored sites primarily were obtained from either Tampa Bay Water or SWFWMD using their applications (DataMart and work management information system [WMIS], respectively). Appropriate care was taken to ensure that a common vertical datum was used, as some records were available in both National Geodetic Vertical Datum (NGVD)29 and North American Vertical Datum (NAVD)88 through WMIS. (An Excel spreadsheet of mean monthly lake levels was provided by Brian Ormiston and Claudia Listopad.) Wetland and lake NPOs were calculated by subtracting HNPs from the median water levels for the period 2008-2014. Medians were calculated using all available data for the period of interest. Visual examination of hydrographs was Continued on page 40

Florida Water Resources Journal â&#x20AC;˘ August 2018


Table 1. Independent variables evaluated for use in predicting wetland/lake median normal pool offsets.

Figure 5. Surficial aquifer system drawdown (ft): 12 nearest-neighbors inverse-distance-squared weighted interpolation.


August 2018 â&#x20AC;˘ Florida Water Resources Journal

Continued from page 38 performed to confirm that the calculated medians would be representative (i.e., excessive dry values might prevent calculation of accurate medians). The HNPs were obtained from a variety of sources that had been compiled for the RA GIS (RAGIS) project (GPI and Applied Ecology, 2015). These HNPs were checked against an environmental management plan (EMP) database maintained by Tampa Bay Water. Specific NPOs are known from past studies (e.g., SWFWMD, 1999; GPI, 2016) to represent recovered conditions for monitored wetlands, depending on the type of wetland (e.g., isolated or connected) and the surrounding soil type (e.g., mesic or xeric). For example, Figure 4 shows that, following the decrease in local groundwater production at the North Pasco Wellfield, the six-year running median water level rose above the surrogate minimum level (calculated as a NPO of -1.8 ft). In other words, when the six-year median is shown to rise above the relevant site-specific threshold NPO, the site is considered â&#x20AC;&#x153;low or no stressâ&#x20AC;? and, therefore, hydrologically recovered. A decision was made to transform the dependent variable, NPO, in order to improve the normality of residuals from the planned linear model. Specifically, the following transformation was used:

Prior to applying the transformation, three sites with very small positive HNPs were adjusted to zero. (One interesting consequence of using a nonlinear transformation on the dependent variable is that prediction intervals generated from a fitted statistical model will not be symmetric when the prediction intervals are transformed back into their original scale.) A variety of independent or auxiliary variables were obtained and prepared to provide the best possible aspatial portion of the RK model. These 12 independent variables are shown in Table 1. Maps of several of the independent variables prepared and investigated (GPI, 2017) are provided here. The SAS DDN (Figure 5) was based on a 12 nearest-neighbors inverse-distance-squared weighted interpolation derived from a point file provided by Tampa Bay Water, representing the maximum of the historical production and scaled pumpage scenarios described (Tampa Bay Water, 2013). In Figure 5, positive numbers represent ft of predicted surficial aquifer DDN in the NTB area associated with anticipated distributions and rates of groundwater production (scaled to 90 mgd). A classified soils layer (Figure 6) prepared from U.S. Department of Agriculture Natural

Resources Conservation Service (USDA-NRCS) soil survey geographic (SSURGO) data in a previous xeric wetlands study (GPI, 2016) was used to determine two variables: xeric ratio and xeric yes/no (Y/N). Xeric ratio was calculated using the methods described in Berryman & Henigar Inc. and SDI Environmental Services Inc. (2000). A 500-ft buffer was constructed around each wetland or lake of interest and the ratio of the following areas calculated:

Xeric Y/Ns representing a binary thresholding of xeric ratios greater than 27 percent were considered “Y” (i.e., yes, they are xeric-associated sites), while those with less than or equal to 27 percent were considered “N” or no. The splitting criterion of 27 percent was based on methods presented in Berryman & Henigar Inc. and SDI Environmental Services Inc. (2000). Mean rainfall for 2008-2014 was calculated based on 11 gap-filled stations using a 10 nearest-neighbors inverse-distance-squared weighted interpolation (Figure 7). All rainfall data were obtained from Tampa Bay Water, except for 26353 CLERMONT 9 S NWS, which was provided by the National Oceanic and Atmospheric Administration (NOAA). The 11 rainfall station locations used were: S RN-CBR-CB01 S RN-STK-STK14 S RN-NOP-NOP S RN-NHW-5 S RN-MBR-3C S RN-NWH-S21 S RN-ELW-METER_PIT S RN-CYB-CYB7 S RN-CYC-CC3 S RN-CNR-T3 S 26353 CLERMONT 9 S NWS

A best subsets regression search was undertaken using the Bayesian information criterion (BIC) to avoid overfitting. Minimization of the BIC allowed identification of the model in a set of candidate models that gave the best balance between model fit and complexity, with the intent that whatever variables resulted in the most probable model from a BIC perspective would also yield valid predictions using out-of-sample data in the future. The best subsets regression, performed using the glmulti package in R (Calcagno and de Mazancourt, 2010), consisted of fitting all possible regression models without interactions (more than 4,000 models) using all

various combinations of the 12 independent variables. The final selected variables were used in a multiple linear regression, allowing interactions (i.e., the final aspatial model). Spatial autocorrelation remaining in the residuals from the final aspatial model was modeled using a variogram as part of the RK process. At a conceptual level, kriging represents a family of interpolation techniques that explicitly model the spatial autocorrelation in the data using a technique called the variogram (Figure 8). The variogram is an empirical measure of how differences between pairs of sample points change Continued on page 42

Figure 6. Classified soils layer from previous xeric wetlands research.

Figure 7. Mean rainfall for 2008-2014 (in.); orange dots represent rain gauges.

Figure 8. Diagram of an idealized variogram (Google, 2016). Florida Water Resources Journal • August 2018


Continued from page 41 with distance. The typical plotted variogram function (the semivariance) rises as distance increases (Figure 8), indicating that the average differences between values at locations in the study area are smaller when the samples are close together and much larger when they are farther apart. The point where the semivariance function flatlines represents the distance (range) at which positive spatial autocorrelation is negligible. Note that a portion of the variance, called the nugget, cannot be explained based on proximity to other samples (Figure 8); it could, however, represent some combination of variability at distances smaller than the typical sample distance, measurement error, or variability due to unmeasured factors. The variogram provides a method for weighting the influence of nearby sites in estimating unknown values (i.e., a datadriven estimate of the spatial covariance between samples). The focus of ecological condition interpolation was the five-year WHA rating. The WHA program has been performed at more than 400 wetlands in the NTB area in four general time periods: 1997/1998 (Rochow, 1998), 2004/2005 (Bureau Veritas et al., 2006), 2009/2010 (GPI Southeast Inc. et al., 2010), and most recently in 2016. Among other data collected at each site, an overall WHA rating on a 1-3 scale is available. This rating represents a “relative estimation of wetland health” with sites rated 1 being considered “severely stressed,” those rated 2 as “moderately stressed,” and those rated 3 as “low or no stress.” Although this rating is ultimately based on the professional opinion of the environmental scientist performing the field evaluation, the opinion is supported by detailed information collected, such as which plant species were pres-

ent at each site and various wetland-quality questions (e.g., “fallen trees”: >25, 5-25, or <5 percent). The WHA ratings do not attribute a cause for observed stress, although relevant observations regarding land use, ditching, etc., are included on the field sheets. Based on availability for this study, a file was prepared representing 2009 conditions to use as the basis for understanding ecological conditions across the study area. The following data were included in the 2009 conditions file: 423 WHA ratings actually collected in 2009/2010 (but 20 of those were outside study area), and 3 ratings only (n=472) from 1997/1998 at those sites where no 2009/2010 data existed. The rationale for including sites rated 3 from 1997/1998 in the 2009 conditions file is that conditions are unlikely to have worsened in most areas due to decreased production. The combined 2009 dataset provided 895 sites with a rating of 1, 2, or 3. In the present study, ecological conditions were interpolated across the NTB area using the inversedistance-squared weighted interpolation in ArcGIS using the 12 nearest neighbors and then assigned to the unmonitored sites.

Hydrologic Results Figure 9 represents the information criteria (IC) profile plot, or the ranked BIC values of the very best models evaluated in the all-subsets regression search through all possible combinations of the 12 candidate auxiliary variables. Note that “best” in this context implies that the lowest BIC value represents the most likely model, minimizing overfitting, given this particular dataset. The red line in Figure 9 is placed two BIC units above the very best model, based on the rule of thumb that any models within two BIC units of the best one are worth considering

Figure 9. Bayesian information criteria profile plot from R package glmulti output.


August 2018 • Florida Water Resources Journal

(Calcagno and de Mazancourt, 2010). When faced with multiple “best” models, investigators may wish to explore a model-averaging approach, also known as multimodel inference (Burnham and Anderson, 1998). In this case, only the very best model was chosen to be implemented (lowest BIC). The best subsets regression search among all possible combinations of the 12 auxiliary variables resulted in just three variables being included in the BIC best model: SAS DDN, xeric ratio, and intermediate aquifer (IA) thickness with a minimum BIC of 392.7 (Figure 9). Although referred to as IA thickness by the creators of the Florida aquifer vulnerability assessment (FAVA) dataset, it’s been noted that much of the NTB area lacks an intermediate aquifer, and this variable might be more appropriately reduced to a simple presence/absence of a confining layer. The final aspatial model selected used the three variables identified by the BIC best subsets search—SAS DDN, xeric ratio, and IA thickness—and allowed for interactions among them. This final model had an adjusted R-squared of 0.33 (Figure 10). Visualizations of the partial residuals were performed using the R package visualization of regression models (visreg) from Breheny and Burchett, 2016. The visualizations documented the modeled effects of each independent variable, while statistically holding other variables constant (either at their median values or two selected values to allow visualization of interactions). The SAS DDN showed the expected relationship, with larger negative NPOs associated with larger DDNs in Figure 11 (larger negative NPOs were represented as larger positive values on the transformed scale). Higher xeric ratios were also associated with larger negative NPOs Continued on page 45

Figure 10. Final aspatial regression model R output.

Florida Water & Pollution Control Operators Association

FWPCOA STATE SHORT SCHOOL August 13 – 17, 2018 Indian River State College - Main Campus – FORT PIERCE –

COURSES Backflow Prevention Assembly Tester ..........................$375/$405

Stormwater Management A .........................................$325/$325

Backflow Prevention Assembly Repairer ......................$275/$305

Utility Customer Relations I, II & III................................$260/$290

Backflow Tester Recertification ......................................$85/$115

Utilities Maintenance III & II ..........................................$225/$255

Basic Electrical and Instrumentation ............................$225/$255

Wastewater Collection System Operator C, B & A ......$325/$325

Facility Management Module I......................................$275/$305

Water Distribution System Operator Level 3, 2 & 1 .....$325/$325

Reclaimed Water Distribution C, B & A ........................$225/$255 (Abbreviated Course) ................................................$125/$155

Wastewater Process Control ........................................$225/$255 Wastewater Troubleshooting ........................................$225/$255

Stormwater Management C & B ...................................$260/$290

For further information on the school, including course registration forms and hotels, visit:

SCHEDULE CHECK-IN: August 12, 2018 1:00 p.m. to 3:00 p.m. CLASSES: Monday – Thursday........8:00 a.m. to 4:30 p.m. Friday........8:00 a.m. to noon

FREE AWARDS LUNCHEON P Wednesday, August 15, 11:30 a.m. P

For more information call the

FWPCOA Training Office 321-383-9690 Florida Water Resources Journal • August 2018


Figure 11. Partial residuals visualization: surficial aquifer system drawdown.

Figure 12. Partial residuals visualization: xeric ratio.

Figure 13. Partial residuals visualization: surficial aquifer system drawdown by xeric ratio.

Figure 14. Partial residuals visualization: surficial aquifer system drawdown by intermediate aquifer thickness.

Figure 15. Experimental residual variogram and variogram model.


August 2018 â&#x20AC;˘ Florida Water Resources Journal

Figure 16. Cross-validated back-transformed regression-kriging residuals.

Continued from page 42 (Figure 12). The relationship between SAS DDN and NPOs showed an interaction with xeric ratio, so that sites with 100 percent xeric ratios tended to have greater negative NPOs at the same magnitude of SAS DDN (Figure 13). The relationship between SAS DDN and NPOs also showed an interaction with IA thickness: sites with IA thickness greater than zero showed greater negative NPOs at the same magnitude of SAS DDN (Figure 14). The RK was used to model spatial autocorrelation in the residuals remaining after the final selected aspatial model. More specifically, an experimental variogram and variogram model were prepared from the residuals from the final aspatial model (Figure 15) using an autofitting procedure available in the R package automap (Hiemstra et al., 2009). The variogram shows a sill with a semivariance of 0.24 and a nugget of 0.12. The implication of this is that about half of the residual semivariance after the regression model may be explained by spatial autocorrelation. If this residual variogram had been just a flat line, the residuals would not have been kriged, as there would be no improvement in prediction over the final aspatial model. As expected, the RK predictions showed a higher correlation with the actual NPOs than the simple aspatial model, as the adjusted R2 for the selected aspatial mode (Figure 10) was about 33 percent. The squared correlation coefficient for the RK approach (i.e., the coefficient of determination) was found to be 52 percent, representing an improvement from the aspatial model; however, the performance on data not used in the development dataset could be worse. To understand future expected performance on unsampled areas, a more realistic estimate of future performance was prepared for the RK model by performing a leave-one-out cross-validation (LOOCV). The LOOCV technique involves refitting the model while leaving out one observation, then deriving an estimate of that one observation as an out-of-sample case. The procedure is performed for all observations in turn, resulting in an estimate of error closer to that expected for a completely novel future dataset (or in this case, hypothetical performance at unmonitored sites located between monitored sites). Although the LOOCV R2 was lower (36 percent) than the overly optimistic model (52 percent), and some predictions were far from the observed data (Figure 16), the majority of LOOCV residuals were within 1 ft. Specifically, 50 percent of the time the RK model is expected to predict median water levels at unmonitored wetlands within a range of 0.70 ft lower than the actual value to 0.88 ft higher than Continued on page 46

Figure 17. Regression-kriging predictions: transformed units.

Figure 18. Regression-kriging predictions: back-transformed to feet.

Figure 19. Regression-kriging standard deviations: transformed units (with overlay of monitored site locations). Florida Water Resources Journal â&#x20AC;˘ August 2018


Figure 20. Regression-kriging predictions for 2008-2014 median wetland water levels transferred to unmonitored sites.

Continued on page 45 the actual value. This level of accuracy is considered useful for wetland recovery planning purposes. The RK predictions for the 2008-2014 median NPOs are presented in map form in trans-

formed (Figure 17) and back-transformed (ft) units (Figure 18). The back-transformed prediction map shows areas in blue predicted to meet both the isolated cypress standard threshold of 1.8 ft below HNP, as well as the xeric wetlands screening criterion of 3.1 ft below HNP (e.g.,

Figure 22. Wetland health assessment data: 2009 12 nearest-neighbors inverse-distance-squared interpolation.


Figure 21. Wetland health assessment data: 2009 conditions as points.

August 2018 â&#x20AC;˘ Florida Water Resources Journal

much of CYB and South Pasco, and portions of Eldridge-Wilde). Other areas shown in yellow represent those areas predicted to meet the xeric wetlands criteria, but not the isolated cypress standard (e.g., Morris Bridge and portions of CYC). Areas shown in orange were predicted to have unmonitored site median water levels lower than 3.1 ft below HNP, including most of Cross Bar Ranch and parts of CYC and Eldridge-Wilde (i.e., areas with water levels not predicted to meet either threshold). A map of RK prediction standard deviations is also provided from the RK analysis (Figure 19). Certain areas had very high uncertainty due to few sample sites nearby (shown in red), including some areas on the Eldridge-Wilde and Cross Bar Ranch wellfields. One of the advantages of the RK approach is that a spatially referenced measure of uncertainty is provided. In future implementations, it may be useful to screen out certain areas of high uncertainty from presentations of results, particularly those areas distant from the wellfields that have very few nearby development dataset observations. When the appropriate thresholds by soil type are considered, overall 253 out of the 684 unmonitored sites (37 percent) are predicted to have met their site-specific thresholds based on RK predictions for the 2008-2014 period (Figure 20). Certain areas showed a high degree of heterogeneity in recovery predictions, such as the areas around the Northwest Hillsborough Wellfield and the southern part of CYB, while other areas tended to be more uniform in showing nearly all unmonitored sites as not recovered

Table 2. Three-group classification matrix for 5 percent held-out sample for 12 nearest-neighbors inverse-distance-weighting interpolation.

Table 3. Two-group classification matrix for 5 percent held-out sample for 12 nearest-neighbors inverse-distance-weighting interpolation.

Figure 23. Interpolated wetland health assessment data from 2009 conditions assigned to unmonitored sites.

(Morris Bridge and much of Cross Bar Ranch) or nearly all as recovered (central CYB). The Eldridge-Wilde Wellfield showed a radial pattern of predicted recovery around the periphery of the wellfield.

Ecological Results Ecological conditions as expressed by the WHA three-point rating for 2009 are presented for 868 wetland points located in the NTB area in the vicinity of the 11 central system Tampa Bay Water wellfields in Figure 21 (with the points representing geographic centroids, constrained to be included within the shape of the original polygon). The use of a relatively simple spatial interpolation algorithm was investigated to reproduce WHA 2009 values for unmonitored wetlands—inverse distance weighting (IDW). The IDW algorithm calculates unknown locations as a weighted average of nearby points, with the weighting function being a function of the distance. Application of a 12 nearest-neighbors-squared IDW algorithm to the points in Figure 21 (WHA 2009 conditions) yielded Figure 22, where certain areas appear to be dominated by severely stressed conditions (orange color), such as western and central ELW, while other areas appear dominated by low- or nostress conditions (green color), such as eastern Starkey (STK) and southern CYB. Some areas in between the wellfields show a speckled appearance, reflecting abrupt transitions in site WHA ratings for sites relatively near each other. In order to evaluate how useful the IDW ap-

proach might be for assessing WHA values at completely unmonitored sites, an out-of-sample evaluation was done. The evaluation was accomplished by randomly selecting 51 of the known data points (approximately 5 percent) to be excluded from an interpolation, and then evaluating how well those missing points could be interpolated from their 12 nearest neighbors. Table 2 is a classification matrix, also referred to as a confusion matrix, representing a comparison of the actual and predicted classifications of the 51 out-of-sample points. The overall accuracy of the classification is based on summing the diagonal elements of the table (i.e., 5, 9, and 18) and dividing by the total count of 51. Therefore, the 12 nearest-neighbors IDW achieves an overall accuracy of 63 percent (i.e., 32 divided by 51). A more useful classification measure, however, would be the ability of the approach to identify sites as either stressed (WHA of 1 or 2) or nonstressed (WHA of 3). The classification matrix for this simpler twogroup case is presented in Table 3. The 12 nearest-neighbors IDW accurately classified the out-of-sample points 71 percent of the time (36 divided by 51). This expected out-of-sample performance of the stressed/nonstressed WHA interpolation of 71 percent represents a useful level of accuracy. As an example, based on the 868 WHA 2009 points occurring in the NTB Area, 327 were considered stressed, for a background rate of stress of 38 percent (327 divided by 868). If there were no other information about an unmonitored site, it could be concluded that there was a 38 percent chance it would be rated stressed in 2009; how-

ever, if the simple 12 nearest-neighbors IDW spatial interpolation model indicates that the site is stressed, the positive predicted value can be calculated as the ratio of true positives (18) divided by the sum of the true positives and the false positives (i.e., the number of predicted positives or 26), or 69 percent. This distance-weighted model has substantially increased the chance of correctly classifying a site as stressed (from a “chance alone” rate of 38 percent to an “interpolated predicted” rate of 69 percent). It’s possible that optimization could be used to further improve the accuracy of the IDW approach. Specifically, there are two primary parameters required by the IDW algorithm (the number of neighbors and the distance-weighting). This example shows the result of using 12 nearest neighbors with a weighting factor of inverse distance squared. Cross-validation could be used to select optimal IDW parameters by iteratively holding out individual known observations and estimating them as “test” observations. Evaluation of the cross-validated errors for these temporary test observations could be used to select minimal test errors out of a matrix of parameters investigated. It’s uncertain the extent to which the IDW model accuracy could be improved through optimization, but it’s unlikely that the values selected by professional opinion are the most optimal, so 71 percent accuracy may be considered the minimum binary classification accuracy achievable through IDW. An alternative to optimizing the IDW parameters would be using an RK approach similar to that used for the hydrologic analysis Continued on page 48

Florida Water Resources Journal • August 2018


Continued from page 47 presented in the previous section, and this is what is intended for future, similar investigations. Interpolated values from the 12 nearestneighbors IDW models were spatially assigned to the unmonitored sites, and the results are presented in Figure 23. Actually, 81 of the 684 unmonitored sites (12 percent) had an existing WHA for 2009, so these ratings superseded the interpolated value. A review of the resulting map suggests that nonstressed unmonitored wetlands were concentrated in south CYB, much of MB and NWH, and portions of eastern ELW and western CBR. Other areas in 2009 showed moderate to severe stress conditions.

Summary and Conclusions A machine-learning approach was effective in screening a large number of variables and developing a spatially explicit estimate of median wetland water levels at unmonitored wetlands and lakes in the NTB area. A hybrid spatial interpolation technique, RK was investigated for interpolating wetland water levels at unmonitored sites. Hybrid spatial interpolation techniques tend to provide more accurate predictions than either individual approach by itself (i.e., regression or kriging). To demonstrate the RK approach, an information theoretic-driven best-subsets regression was first used to develop the best possible aspatial regression model to predict historical median (2008-2014) NPOs at 309 monitored sites with appropriate data. Three variables useful for predicting water levels were selected from the 12 evaluated: SAS DDN, xeric ratio, and IA thickness (i.e., presence/absence of confining unit). Once the best aspatial regression model was developed, residuals from the model were kriged to statistically characterize their spatial autocorrelation. A cross-validation approach was used to assess the likely performance of the RK model on future datasets. Performance of the final model was considered useful, with cross-validated residuals indicating that at least half the time the combined RK model is expected to predict water levels at unmonitored wetlands within a range of 0.70 ft lower than the actual value to 0.88 ft higher than the actual value. When the appropriate thresholds by soil type are considered, overall 253 out of the 684 unmonitored sites (37 percent) met the appropriate hydrologic metric, based on RK predictions for the 2008-2014 period. Certain areas showed a high degree of heterogeneity in median wetland water-level predictions, such as the areas around the Northwest Hillsborough Wellfield and the southern part of CYB, while other


areas tended to be more uniform in showing nearly all unmonitored sites as not meeting the metric (Morris Bridge and much of Cross Bar Ranch), or nearly all as meeting the established metric (central CYB). The Eldridge-Wilde Wellfield showed a radial pattern of predicted recovery around the periphery of the wellfield. The RK framework provides the flexibility to include other data-mining algorithms in the “regression” portion. For example, Appelhans et al. (2015) studied 14 machine-learning algorithms and kriging for interpolating air temperature on Mt. Kilimanjaro and found that numerous tree-based modern data-mining algorithms outperformed both linear regression models and universal kriging, including stochastic gradient boosting, cubist, and random forest. Ultimately, the authors elected to use an RK framework with the cubist model in place of regression (Appelhans et al., 2015). It was recommended that one or more modern datamining algorithms be tested in a future study for comparison to the multiple linear regression approach used in this study. Although prediction using a different algorithm may be improved, it’s likely that interpretability of the effects of the various auxiliary variables will suffer, but prediction is more important than interpretability for the application of estimating NPOs at the unmonitored sites. If multiple predictions are available from independent techniques, each with their own measure of error, they may be combined by weighting their predictions for each cell by the inverse of their errors (Hengl, 2009). Intuitively, diverse algorithms that perform relatively poorly in different geographic areas could be combined to yield overall improved predictions throughout the region of interest. The five-year WHA program provides a spatially and temporally rich dataset able to support interpolation of wetland conditions, and potentially, changes in conditions over time. The WHA scores representing 2009 conditions documented at 868 sites (and including nonstressed sites from 1997/1998) were used with an IDW algorithm to successfully interpolate ecological conditions at unmonitored sites. Using an outof-sample analysis, the distance-weighted model substantially increased the chance of correctly classifying a site as stressed (from a “chance alone” rate of 38 percent to an “interpolated predicted” rate of 69 percent). The availability of multiple sampling events separated by approximate five-year periods also would allow a metric calculation of ecological change in a future study. An update of the analysis presented here will be done to include data collected in 2016, which are expected to be more representative of postcutback eco-

August 2018 • Florida Water Resources Journal

logical conditions because certain wellfields (e.g., Starkey) were not able to reduce groundwater production until the end of 2007. In addition, a time lag is anticipated between the time of production cutbacks and ecological recovery, so the additional years between the 2009 conditions and 2016 conditions are expected to reveal a more nearest-neighbor accurate picture of the extent of recovery. It’s anticipated that both hydrologic and ecological condition interpolations will provide valuable evidence for assigning unmonitored sites to appropriate RA status bins. It's recommended that geostatistical predictions of hydrologic parameters be given precedence over ecologic predictions because hydrologic changes may precede ecological changes in wetlands, and hydrologic data measurements are inherently more precise than ecological WHA ratings. Geostatistical predictions of NPO may be used to quickly identify those sites believed to be “recovered,” or meeting their site-specific median water level threshold. Sites predicted to not meet their site-specific thresholds will be categorized as either “improved, not fully recovered” (INFR), “not fully recovered, continuing wellfield impact” (NFRC), or impacted due to other causes (e.g., surface drainage alterations). Distinguishing between the first two cases may be primarily based on how far below predicted water levels are relative to the HNP, or possibly, the wetland bottom. (INFR sites might be near their site-specific thresholds, while NFRC sites are expected to be still relatively far below their thresholds.) The WHA predictions (and predicted changes in WHA conditions) are expected to provide additional evidence to guide categorization. The results of this study represent a “proof of concept” only. The results suggest RK provides a useful method for recovery analyses of unmonitored wetlands in the area of investigation. The study will be repeated with updated datasets and using other machine-learning approaches (such as random forest). In addition, Tampa Bay Water is using a “weight of evidence” approach to recovery analysis. The RK results will be one parameter considered in the final assessment of recovery status; other hydrological and ecological analyses and professional opinions will factor in the final decisions regarding recovery status. An attempt will be made to assess the relative accuracy of the RK predictions for wetlands in developed landscapes (e.g., suburban areas), as the water budgets of these wetlands may be affected by surface drainage or land-use influences not reflected in the RK model. It’s recommend that the more time-intensive tools available to assess unmonitored sites—aerial imagery analysis and field visits—

be reserved for validating assignments based on the statistical binning process; that is, representative sites may be visited to confirm, for example, that recent hydrological and ecological field indicators point to a particular group of sites being below their threshold, but making sufficient improvement to be considered INFR. In addition to visiting representative sites of welldefined groups, it’s likely there will be some unusual or uncertain sites that require field visits to confirm unusual circumstances, such as suspected drainage alterations, excessive soil subsidence, or other site-specific factors. In conclusion, to answer the question posed by this article’s title, it’s believed that machine-learning techniques, combined with geostatistical methods, show great promise for assessing recovery of water levels and ecological conditions at unmonitored wetlands and lakes. More broadly, the application of a hybrid statistical method can allow water managers to make better predictions by accounting for known variables believed to influence water levels, as well as unknown, spatially autocorrelated factors.

References • Alpaydin, E. 2009. Introduction to Machine Learning, Second Edition. MIT Press. • Applehans, T., E. Mwangomo, D.R. Hardy, A. Hempd, T. Naussa. 2015. Evaluating machine learning approaches for the interpolation of monthly air temperature at Mt. Kilimanjaro, Tanzania. Spatial Statistics 14:91-113. • Arthur, J.D., A.E. Baker, J.R. Cichon, A.R. Wood, and A. Rudin. 2005. Florida Aquifer Vulnerability Assessment (FAVA) Contamination Potential of Florida’s Principal Aquifer Systems. A report submitted to the Division of Water Resource Management Florida Department of Environmental Protection. < ep_report/FAVA_REPORT_MASTER_DOC_3 -21-05.pdf> Accessed 12/7/2016. • Bellino, Jason C. 2011. Digital surfaces and hydrogeologic data for the Floridan aquifer system in Florida and in parts of Georgia, Alabama, and South Carolina. U.S. Geological Survey Data Series 584. <> Accessed 12/7/2016. • Berryman & Henigar and SDI Environmental Services. 2000. Candidate Sites Evaluation Study. Final report to Tampa Bay Water. Clearwater, Fla. • Breheny, P. and W. Burchett. 2016. Visualization of Regression Models using visreg. <> Accessed 12/12/2016. • Bureau Veritas/Berryman & Henigar, PBS&J,

and EHI. 2006. Five Year Wetland Assessment. Final report to the Southwest Florida Water Management District. Burnham, K.P. and D.R. Anderson. 2002. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer-Verlag. New York. Calcagno, V. and C. de Mazancourt. 2010. glmulti: An R Package for Easy Automated Model Selection with (Generalized) Linear Models. Journal of Statistical Software 34(12):1-29. Desbarats, A.J., C.E. Logan, M.J. Hinton, and D.R. Sharpe. 2002. On the kriging of water table elevations using collateral information from a digital elevation model. Journal of Hydrology 255:25-38. GPI (Greenman-Pedersen, Inc.). 2017. Development of Proposed Methods for Recovery Assessment Bin Assignment of Unmonitored Sites. Final Report to Tampa Bay Water. GPI (Greenman-Pedersen, Inc.). 2016. Development of a Water Level Recovery Metric for Xeric-associated Wetlands in the Northern Tampa Bay Area. Final Report for Tampa Bay Water. GPI (Greenman-Pedersen, Inc.). 2015. Recovery Assessment Geodatabase: Needs Assessment, Draft Schema, and Recommendations. Final Report to Tampa Bay Water. GPI Southeast, Inc., University of South Florida, and Vanasse, Hangen, Brustlin, Inc. 2010. Five Year Wetland Assessment. Draft report to the Southwest Florida Water Management District. Google. 2016. (Diagram of idealized variogram.) Copied without changes from terpolation (accessed 1/15/2017) under Creative commons 3.0 Unported license ( James, G., D. Witten, T. Hastie, and R. Tibshirani. 2017. An Introduction to Statistical Learning with Applications in R. Springer Science+Business Media. On-line resource, retrieved September 27, 2017, from Printing.pdf Hengl, T. G.B.M. Heuvelink, B. Kempen, J.G.B. Leenaars, M.G. Walsh, K.D. shepherd, A. Sila, R.A. MacMillan, J. Mendes de Jesus, L. Tamene, J. E. Tondoh. 2015. Mapping Soil Properties of Africa at 250 m Resolution: Random Forests Significantly Improve Current Predictions. PLoS One 10(6): e0125814. Published online 2015 Jun 25. doi: Hengl, T. 2009. A Practical Guide to Geostatistical Mapping. Published electronically and available under the Creative Commons Attri-

bution-Noncommercial-No Derivative Works 3.0 license: (Accessed 1/11/2017). Hengl, T. G.B.M. Heuvelink, D.G. Rossiter. 2007. About regression-kriging: From equations to case studies. Computers & Geosciences 33:1301-1315. Hengl, T., G.B.M. Heuvelink, A. Stein. 2003. Comparison of kriging with external drift and regression-kriging. Technical note, ITC, Available on-line at Hiemstra , P.H., E.J. Pebesma, C.J.W. Twenhofel, G.B.M. Heuvelink. 2009. Real-time automatic interpolation of ambient gamma dose rates from the Dutch Radioactivity Monitoring Network. Computers & Geosciences 35(8):1711-1721. Lee, T.M., and Fouad, G.G. 2014. Creating a monthly time series of the potentiometric surface in the Upper Floridan aquifer, Northern Tampa Bay area, Florida, January 2000–December 2009: U.S. Geological Survey Scientific Investigations Report 2014–5038, 26 p., <> Accessed 12/7/2016. ISSN 2328-0328. Odeh, I.O.A., A.B. McBratney, D.J. Chittleborough. 1994. Spatial prediction of soil properties from landform attributes derived from a digital elevation model. Geoderma 63:197-214. Pardo-Iguzquiza, E. 1998. Comparison of geostatistical methods for estimating the areal average climatological rainfall mean using data on precipitation and topography. International Journal of Climatology 18:1031-1047. Rochow, T.F. 1998. The Effects of Water Table Level Changes on Fresh-water Marsh and Cypress Wetlands in the Northern Tampa Bay Region: A Review. Environmental Section Technical Report 1998-1. Prepared by the Southwest Florida Water Management District. Brooksville, Fla. SWFWMD (Southwest Florida Water Management District). 1999. Northern Tampa Bay Minimum Flows & Levels White Papers. Southwest Florida Water Management District. Brooksville, Fla. Tampa Bay Water. 2017. Recovery Assessment. On-line resource, retrieved September 27, 2017, from Tampa Bay Water. 2013. Defining Areas of Investigation for Recovery Analysis. Memorandum to the Southwest Florida Water Management District. Clearwater, Fla. Tampa Bay Water. 2012. Permit Recovery Assessment Work Plan and Schedule. Memorandum submitted to the Southwest Florida Water Management District. Clearwater, Fla. S

Florida Water Resources Journal • August 2018



The Owner’s Role in a Construction Project It’s the Money Lauren C. Atwell

isk allocation is a fundamental consideration in all contracts. Usually, parties to construction contracts will seek to include provisions that limit and distribute their respective risks, duties, responsibilities, and liabilities. Such provisions can be unreasonable in their attempts to shift responsibilities from one party to another, and, therefore, should be scrutinized very closely during contract formation and negotiation. Even where express contract provisions are provided, the law generally imposes implied warranties, duties, and responsibilities on the parties. Besides writing checks, the owner’s duties and responsibilities include warranting the plans and specifications, providing site surveys, disclosing superior knowledge, acting on clarifications and changes, interpreting the documents, and cooperating with the contractor. While each construction project has its unique requirements, some obligations typical in most projects are discussed.



Above all, the owner must assure adequate funding for the project to be paid in a timely manner. In addition, part of that task is to arrange reserve funding for the inevitable problems that arise in any job. Some would list this as the owner's first responsibility to the contractor, and it’s probably the primary area of disputes between an owner and a contractor. In the case of a public bid, if the funding source had not been properly secured, it’s at best unfair to those contractors who had taken the time and expended the energy and money to bid the project. Once the project is underway, one of the owner's primary responsibilities, as far as the contractor is concerned, is to secure funds to allow payments as prescribed in the contract. Change orders are a normal part of the construction process. As such, the owner has an obligation to anticipate their occurrence. Project funding should incorporate some additional percentage (often 5 to 7 percent) of the project bid amount to be budgeted to fund legitimate changes as they occur. Responsibility for the delay in the owner’s response to proper contract changes that results from lack of, or late availability of, funds will often rest with the

August 2018 • Florida Water Resources Journal

owner. Change orders occupy the single greatest source of litigation/arbitration in construction projects.

Providing Site Surveys It’s the owner's responsibility to provide complete, accurate, and relevant data, as it may become necessary for correct installation of the work. The contractor is typically responsible for the correct layout and execution of the work. If excavation is required, geotechnical data describing the soil composition that will have to be dealt with is a fundamental prerequisite. The interesting corollary here is that the boring and soils information should be given in locations properly relevant to the construction. For instance, if the boring information around the perimeter of a foundation indicates sand with a low water table, and no information regarding the soils characteristics is given within the foundation area, the contractor may assume sand with no water in the way of the excavation exists throughout the entire foundation construction. When either clay or silty soil is subsequently encountered while excavating the interior portions of the foundation (where no specific information had been given), the contractor may be entitled to additional compensation. The owner is typically responsible for providing accurate locations of all existing utilities. Locations of fiber optic lines may be necessary to tie in new building services or prevent accidental interruption. Correct sanitary and storm line locations and elevations are critical to the design of underground drainage systems and tie-ins. Invert elevations are also necessary to allow proper estimates of the amount of excavation and backfill for the respective lines. If, for example, the invert elevation indicated on the drawings as -4.5 feet is actually -9.5 feet, the additional 5 feet of excavation depth may require shoring or greater trench width—all at increased cost to the contractor. If this is due to an improper representation on the drawings of the subject utility location, the owner will often pay for the increased costs. The relationship of adjacent properties may be significant to the construction on the site. An important piece of information that may not be disclosed in the contract documents and may not be readily apparent in a prebid site investigation by the contractor, for instance, is

the drainage characteristic of a surrounding property. For example, if surface water from several acres of land drains into a swale that creates an active watercourse through the site that lasts several days each time it rains, the responsibility for resulting downtime, additional drainage requirements, and reworking of the affected areas will likely rest with the owner.

Warranting the Plans and Specifications The owner usually warrants the adequacy of the plans and specifications on many projects and, therefore, bears the responsibility for any defects or deficiencies in them. Such defects can exist in many forms, but tend to fall into two categories: product and time. Most defective specification problems involve the inaccuracy of the technical specifications. Time most often becomes a consequential consideration related to the failure of the contract to provide the technical requirements accurately and completely. When the owner sole-sources material and/or equipment to the contractor for use in the work, there is a warranty that these items will be suitable for their intended purpose.

Disclosing Superior Knowledge The owner has a duty to disclose superior knowledge to the contractor that may directly or indirectly relate to the work, where that knowledge is either unknown by, or has not been made available to, the contractor. Referring back to the boring data case example, assume that the boring and soils data did in fact exist for the interior portions of the foundation construction, but the data were intentionally left out of the contract. The hope that the contractor might absorb the cost of working through the clay and water (once it’s finally encountered) that would have otherwise been disclosed by the proper inclusion of the relevant information may leave the owner with liability. Likewise, where the owner's superior knowledge of a factor, such as the unavailability or inadequacy of a specified material, would lead to reduced costs, improved efficiency, or simply an earlier disclosure of a problem, the owner has an obligation to advise the contractor.

Action on Clarifications and Changes Construction contracts, whether produced by and for a private owner or public agency, recognize the need and importance of timeliness with respect to issuing clarifications and reviewing change orders. This is an acknowledgment that change orders can interfere with and

disrupt the orderly sequence of the work, and that they should be resolved as quickly as possible to minimize their potential direct and consequential impact on construction. The process of anticipating, identifying, pricing, negotiating, and executing changes, change orders, and change directives is perhaps the most prominent thread through most construction disputes. They are inevitable and controversial, and even if the contract provides for clear processes, most courts and arbitrators will ignore any terms that strike them as “unfair.” Above all, regarding areas in the owner’s realm of responsibility, change orders are the ones that require the most attention and care.

Interpretation of the Documents Although the engineer normally researches and prepares recommendations for technical matters relating to design, when involving quality, cost, and/or time, these corrections, changes, and interpretations are communicated to the owner, who often controls the final disposition on the matter.

Cooperating with the Contractor Even when not expressed in the contract, it’s generally understood that the owner has an implied duty to cooperate with the contractor to the best of his or her ability, and not to impede, hinder, obstruct, or interfere with the work. The same concept applies to the contractor. Although this sounds simple, its success or failure often boils down to the personalities involved. Both parties are focused on the traditional objectives of time, cost, and budget, but to differing degrees and from different perspectives. A consistent theme in disputes is the disintegration of cooperation because of conflicts between owner and contractor over the achievement of time, cost, and quality objectives. Building life-changing infrastructure that provides clean water for all people that will last for generations can be one of the most satisfying of all occupations; however, to counter that creative and often profitable advantage is the need to take risks and master an entire galaxy of skills. The items discussed here are only the beginning. Getting experienced contractors is necessary for the utility owner, and retaining a flexible mind, keeping alert to the innumerable problems that inevitably arise—and keeping a sense of humor—are all central mental preparations for this remarkable challenge. Lauren C. Atwell is chief operating officer with Petticoat-Schmitt Civil Contractors Inc. in Jacksonville. S

News Beat Chris Owen recently joined Hazen and Sawyer as director of water and reuse innovation, where her focus will be drinking water and reuse research and supporting the “One Water” research strategy. She has worked in the industry for more than 27 years. Prior to joining Hazen, she worked for Tampa Bay Water for 16 years in varying capacities, including water quality senior manager, regulatory compliance senior manager, and water quality assurance officer. She was responsible for permitting and compliance, the laboratory, source water assessment, and applied water treatment research. Owen has served on the EPA Science Advisory Board Drinking Water Committee and the EPA National Advisory Council for Environmental Policy and Technology Subcommittee for Technology and Policy. She is a trustee for the AWWA Water Science and Research Division, serves on the board of directors for the American Membrane Technology Association (AMTA), and was named the 2017 Water Quality Person of the Year by AWWA and AMTA.


Gresham, Smith and Partners (GS&P) was recently selected by Pinellas County to provide engineering services for upgrades to the biosolids facilities at the South Cross Bayou Water Reclamation Facility. The facility is the county's largest wastewater treatment facility and distributes an average of 33 mgd of reclaimed wastewater to residents and businesses, while also processing all of the county's biosolids. The firm will evaluate current conditions and design a state-of-the-art biosolids dewatering system to enhance resource recovery and support the county's sustainability goals. "By treating water and recycling biosolids, we can complete a natural cycle in the environment rather than taking up space in a landfill," said Jody Barksdale, P.E., ENV SP, senior vice president and biosolids practice leader with GS&P. "Pinellas County is dedicated to sustainable resource recovery and we look forward to supporting its commitment. Continued on page 53

Florida Water Resources Journal • August 2018


WEFTEC® Opening General Session Speaker Highlights How to Be a Water Sector Hero he water sector is full of everyday heroes. At WEFTEC® 2018 in New Orleans, the opening general session keynote speaker, Kevin Brown, will share his vision for increasing the ranks. Brown, a motivational speaker and author of the book, The HERO Effect, will share his ideas, strategies, and principles to inspire water professionals to recognize and embrace being the everyday heroes who show up and give their best when it matters the most. In this question-and-answer article, Brown shared aspects of his message, as well as his hopes for what WEFTEC attendees will take away from his talk.


Your speech during the opening general session will focus on your book, “The Hero Effect.” Can you summarize exactly what you mean by that and how it differs from the conventional understanding of being a hero? The HERO Effect is a shift in the conventional way of thinking about heroes. Traditional thinking suggests that heroes are ordinary people who do extraordinary things. What I’ve discovered on this journey is that heroes and high performers are anything but ordinary. Ordinary is a learned behavior. Think about this: If heroes are ordinary people doing extraordinary things, then by default we give ourselves permission to be ordinary most of the time, with only the occasional burst of extraordinary. The HERO Effect is about what happens when extraordinary people choose not to be ordinary. The HERO Effect is comprised of four principles: S Heroes help people, with no strings attached. They go “all in” every time they take the field. They understand that, in business and in life,


Kevin Brown

it’s always personal and never perfect. Heroes create strong connections and reach beyond the borders of transactional thinking to create transformational moments! S Heroes create an exceptional experience; the hero’s calling card is pure excellence. It’s about using talents, gifts, and abilities to their fullest potential and highest purpose. Heroes uses the best of who they are to serve more people, more often, and in bigger and better ways. Heroes are committed to personal development and show up better today than they were yesterday. A hero builds trust by serving others with an authentic passion that turns everyday moments into superhuman experiences. The hero is driven to serve others and understands that the greatest rewards in life are determined by how well we take care of the people we live and work with. S Heroes take responsibility; they own the moments that matter. They are actively present and engaged and do not believe in random acts of kindness; instead, the hero is motivated by intentional acts of “difference making.” They live by a simple code: “Bring your best stuff to the present moment and pour it into the lives of others.” They understand that before you can

August 2018 • Florida Water Resources Journal

lead anyone else, you must first be able to lead yourself. Heroes own their attitude, actions, and results. They are committed to the best possible outcome in every situation, regardless of circumstances or events beyond their control. The hero leads by example and knows that true success is found in the power of simple choices. S Heroes live and work with optimism; they see the world differently. For them, it’s not about positive thinking, but about perspective. Looking through the lens of optimism gives heroes supernatural vision—they see what others cannot. They see opportunities instead of obstacles and possibilities instead of problems. When things go wrong (and they will) optimism is what helps the hero turn life’s messes into a masterpiece. Participants will learn how to achieve greater results by eliminating “ordinary” thinking and mastering the habit of excellence. I’ll show how they can own the moments that matter (and they all matter) by taking responsibility for their attitude, their actions, and their results, and how to create meaningful relationships and deliver an extraordinary experience for every “customer” at work and at home. What is your background and how does it relate to you becoming a hero? My life is truly the hero’s journey. I wouldn’t be here today if not for the heroes that poured themselves unselfishly into my life, moving me from a place of desperation to a place of inspiration. My gift back to them is becoming the best version of myself and bringing that to the people I live and work with every day. That is everyone’s responsibility in life. I had a pretty unconventional path to where I am today. I grew up in Muskegon, Mich., where my blue-collar roots taught me the value of hard work and determination. My resume includes an eclectic mix of career stops that ultimately led me to the purchase of a franchise at the age of seventeen.

With a lot of help from mentors and friends, I was able to work my way from the front lines in business to the executive boardroom. For nearly two decades I was a sales and marketing executive that helped grow a littleknown family business, SERVPRO (in Gallatin, Tenn.), a cleanup and restoration company, into the number one brand in the industry with annual revenues reaching $2 billion. After a career in franchising that spanned more than 30 years, I decided to retire from corporate America and pursue my passion for bringing my message to as many people and organizations as possible. Were you able to apply what you learned in your own professional life at SERVPRO? As a leader, I believe we have an obligation to the people around us to create an environment where people can be the best version of themselves. My passion is to embrace a simple philosophy that separates world-class organizations and high-performance people from everybody else. That philosophy is The HERO Effect. My great obsession in business and in life is to help people expand their vision, develop their potential, and grow their results. And, as the father of a child with autism, I know firsthand how the principles of true success reach beyond the boardroom and into the lives of real people facing the challenges of everyday life. The HERO Effect transformed me as a leader, father, husband, friend, and fellow human being. It completely changed the trajectory of my career and embodies the essence of true success and genuine leadership. What are the usual obstacles, or “kryptonite” as you term it in your book, that people encounter in becoming a hero? How do they overcome these obstacles? First of all, being a hero is about being your best when it matters the most. The gold standard for what it means to be a hero can be found in our military men and women, first responders, and people from all walks of life who operate at a higher level. Being your best when it matters the most requires a commitment to yourself and the people around you. It takes work. It’s easy to be your best when everything is going well, but when the tough times come, that’s when it gets hard. To be your best when you don’t feel like it. To be your best when others may or may not deserve it. To be your best when you’re tired, sick, or frustrated. Heroes and high performers always show up and deliver the goods. Always. The single most destructive force on the

planet is a decision to be ordinary; to deprive the world of all that you were put here to be and to rob yourself and the people around you from your greatest contribution. Think about it: Superman spent most of his time disguised as an ordinary person, and as a child I remember thinking, if you could be Superman all the time why would you ever choose to be Clark Kent? Are there any famous examples that you would say are a good representation of a hero, as you would define it? Our military men and women and first responders. The world changers, like Nelson Mandela, Martin Luther King Jr., and Mother Teresa. You can look around and see examples in every area of endeavor: business, entertainment, communities, and families. It’s the people and organizations that we pull out of the pile and separate from everyone else. We deem them as different—special in fact. When they show up to do what they do, they operate at a level most people don’t even aspire to. They show up every time with their best stuff and pour it into the lives of others. Why do you think it’s important for those involved in the water sector to understand your message? How can they apply it in their respective fields? Because if they don’t bring their best when it matters the most, bad things happen. The communities and people they serve get hurt. Every living thing on the face of this earth depends on its ability to execute at a high level; to create, innovate, and find solutions to problems that impact our most precious and powerful resource. How can they apply it? That’s a great question and I cannot wait to answer that during my keynote address. We are going to have a great time and I assure you I am unlike any speaker they have ever heard. I will deliver a life message with powerful business implications. What do you hope attendees at WEFTEC take away from your talk? My most sincere desire is that they will see the heroes around them and develop the one within themselves. If ever there was a time that our world needed heroes at work and at home, it’s now. The ideas I am going to share have changed the course of my life forever and I can’t wait to share them with all of you.

News Beat

Continued from page 51 Integrating the latest technology into the facility will create a value-added resource recovery process for the county, offsetting energy usage and stretching ratepayer dollars." As an engineering consultant, GS&P will design improvements to three highspeed dewatering centrifuges, dewatered conveyance equipment, polymer storage and feed systems, and electrical and instrumentation equipment. The project team will also conduct a sustainability review of the facility's overall solids processes to verify that energy efficiency goals are met and identify areas to implement future sustainable design features. The water and environment team consists of experienced water and wastewater process experts, engineers, designers, operational and regulatory specialists, and environmental scientists providing a wide range of water, wastewater, and stormwater services. By staying abreast of the latest developments in technology, innovation and regulatory legislation, the team integrates these services with other in-house capabilities to maximize project speed, efficiency, and cost-effectiveness.


Stantec has opened a new office in Florida’s Space Coast. The global design firm’s new location is in Melbourne, which is Stantec’s twenty-second location in the state. The firm’s more than 720 staff provide services in coastal, civil, transportation, water, and electrical engineering; planning and landscape architecture; architecture and interior design; program and project management; environmental services; and construction engineering and inspection. Fermin A. Diaz, P.E., Stantec’s gulf region leader, said, “We’ve established the Melbourne office in direct response to our local clients and community needs. This new location will allow us to better serve our existing and future clients, and provide a local presence, which is always one of our highest priorities.” S

Kevin Brown is a motivational speaker and author of the book, The HERO Effect. He will deliver the keynote address for the opening general session at WEFTEC® 2018 in New Orleans on S October 1. (photo: Kevin Brown) Florida Water Resources Journal • August 2018


FWEA COMMITTEE CORNER Welcome to the FWEA Committee Corner! The Member Relations Committee of the Florida Water Environment Association hosts this article to celebrate the success of recent association chapter activities and inform members of upcoming events. To have information included for your chapter, send details to Megan Nelson at

Infiltration and Inflow From Cradle to Grave Jamison Tondreault The FWEA Collection System Committee recently hosted a workshop at the Florida Water Resources Conference (FWRC) in Daytona Beach. The workshop, “Infiltration and Inflow from Cradle to Grave,” covered infiltration and inflow (I&I) issues that collection systems typically experience, along with steps taken to help mitigate the issues. Topics at the workshop included: S Multiconsultant Approach to Flow Metering in Pinellas County S Manatee County’s Quantification Approach to Prioritizing Collection System and Infiltration and Inflow Rehabilitation S Removing Private Property Infiltration and Inflow S Lessons Learned from Several Consent Decree Programs

reviewing supervisory control and data acquisition (SCADA) information. Private I&I was the focus of the workshop and multiple ways to reduce it were discussed, including: 1. Installing lockable cleanout covers. 2. Smoke testing the lateral lines and notifying the customers of any leakage. 3. Replacing lateral lines within a public right of way if a leak is detected. 4. Rehabbing lateral liens through lining/lateral grouting. 5. Having an ordinance in place to inspect homes. 6. Installing flow meters in a manhole receiving only flow from customers and notifying customers if spikes in flows are received.

For information on how to get involved with the Collection System Committee, please contact Joan Fernandez (Joan.I.Fernandez or myself (Jamison.Tondreault

During the roundtable discussion, some of the biggest problems and obstacles with combatting private I&I included not enough employees to inspect the laterals, inadequate funds

Jamison Tondreault, P.E., is a water/wastewater engineer with Kimley-Horn & Associates Inc. in Lakeland. S

A slide on how to detect and reduce infiltration and inflow.

Audience members listen to the roundtable discussion.

The workshop concluded with a roundtable discussion of private I&I with esteemed panelists from Miami-Dade Water and Sewer Department, Hillsborough County, Manatee County, Pinellas County, and the City of Lakeland. All presentations concluded that I&I is an important topic and there are multiple ways to detect, quantify, and reduce it. The consensus on detecting and quantifying I&I included smoke testing, monitoring the flows, and calculating the spike in flows during and after a rainfall event. Flows were monitored utilizing flow meters and/or

Roundtable discussion on private infiltration and inflow.


to rehabilitate and replace lines, and putting an ordinance in place to force customers to fix their I&I issues. Special thanks to our volunteers: S Joan Fernandez, who moderated the workshop S Nick Wagner, who helped set up the workshop S Our speakers Jeremy Waugh, Hal Humphrey, Laura Baumberger, and Tony Dill S Our roundtable panelists Juan Bedoya, Richard Cummings, Nickolas Wagner, Richard Ruede, and Kevin Becotte

August 2018 • Florida Water Resources Journal

Speaker Laura Baumberger starts the workshop off with her presentation, “Manatee County’s Quantification Approach to Prioritizing Collection System Infiltration and Inflow Rehabilitation.”

New Products To meet conductivity control needs in online water quality and process applications, Sensorex has introduced the SensoPro Toroidal Conductivity Monitoring System. The system combines a Sensorex TCS3020 probe, capable of reliable conductivity measurements in even the harshest of environments, with the company’s new EX2000RS transmitter, featuring Modbus communication for robust system integration. Monitoring with SensoPro can prevent scaling and corrosion, reduce excessive water usage, and optimize processes in a range of applications, including cooling tower water control, wastewater treatment, brine analysis, desalination, chemical processing, and other harsh or high-conductivity environments. The TCS3020 probe used in the system measures conductivity using toroidal sensing technology, which is more stable and reliable compared to traditional contacting conductivity sensors. Toroidal conductivity sensors do not cause polarization or become fouled and rarely require maintenance. This design, along with the TCS3020’s rugged Noryl-body, reduces maintenance and increases reliability, resulting in more cost-effective operation over time than contacting sensors. The Modbus-equipped EX2000RS toroidal conductivity transmitter completes the system. The unit includes one analog output and can monitor conductivity, percent concentration, total dissolved solids, and salinity. The RS-485 serial interface with Modbus RTU or ASCII protocol provides multiple output options for integration with industrial automation systems. The onboard screen displays measurements in real time and is capable of showing up to four weeks of historical measured values. The EX2000RS has a small-form factor, an IP65/Nema4X case, weighs just 1.1 pounds, and can be wall- or panelmounted. The system is backed by a one-year limited warranty. As a whole, the system features a conductivity measuring range of 0.0 µS/cm-2000 mS/cm, percent concentration measurements for NaCl, HCI, HNO3, NaOH, H2SO4, H3PO4, and automatic or manual temperature control. The result is a reliable conductivity and concentration monitoring system in one simple package. ( S Florida Water Resources Journal • August 2018



Tank Engineering And Management

Consultants, Inc.

Engineering • Inspection Aboveground Storage Tank Specialists Mulberry, Florida • Since 1983





Motor & Utility Services, LLC

Showcase Your Company in the Engineering or Equipment & Services Directory Contact Mike Delaney at


CEC Motor & Utility Services, LLC 1751 12th Street East Palmetto, FL. 34221 Phone - 941-845-1030 Fax – 941-845-1049 • Motor & Pump Services Test Loaded up to 4000HP, 4160-Volts • Premier Distributor for Worldwide Hyundai Motors up to 35,000HP • Specialists in rebuilding motors, pumps, blowers, & drives • UL 508A Panel Shop, engineer/design/build/install/commission • Lift Station Rehabilitation Services, GC License # CGC1520078 • Predictive Maintenance Services, vibration, IR, oil sampling • Authorized Sales & Service for Aurora Vertical Hollow Shaft Motors

CLASSIFIEDS CLASSIFIED ADVERTISING RATES - Classified ads are $20 per line for a 60 character line (including spaces and punctuation), $60 minimum. The price includes publication in both the magazine and our Web site. Short positions wanted ads are run one time for no charge and are subject to editing.

P os i ti on s Ava i l a b l e CITY OF WINTER GARDEN – POSITIONS AVAILABLE The City of Winter Garden is currently accepting applications for the following positions: - Wastewater Plant Operator – Trainee - Solid Waste Worker I, II & III - Collection Field Tech – I, II, & III - Distribution Field Tech – I, II, & III - Public Service Worker II - Stormwater Please visit our website at for complete job descriptions and to apply. Applications may be submitted online, in person or faxed to 407-877-2795.

Reiss Engineering delivers highly technical water and wastewater planning, design, and construction management services for public agencies throughout Florida. Reiss Engineering is seeking top-notch talent to join our team! Available Positions Include: Business Development Leader – Tampa Area Client Services Manager Water Process Discipline Leader Senior Water/Wastewater Project Manager Wastewater Process Senior Engineer Project Engineer (Multiple Openings, 0-15 yrs. exp.)

Engineering Inspector II & Senior Engineering Inspector

To view position details and submit your resume:

Involves highly technical work in the field of civil engineering construction inspection including responsibility for inspecting a variety of construction projects for conformance with engineering plans and specifications. Projects involve roadways, stormwater facilities, portable water distribution systems, sanitary pump stations, gravity sewer collection systems, reclaimed water distribution systems, portable water treatment and wastewater treatment facilities. Salary is DOQ. The City of Winter Garden is an EOE/DFWP that encourages and promotes a diverse workforce. Please apply at

City of St. Petersburg Designer II (Wastewater Utilities) (IRC43631)

Position Requirements: Possession of the following or the ability to obtain within 6 months of hire: (1) Florida Department of Environmental Protection (FDEP) Stormwater Certification and an (2) Orange County Underground Utility Competency Card. A valid Florida Driver’s License is required. • Inspector II: High School Diploma or equivalent and 7 years of progressively responsible experience in construction inspection or testing of capital improvement and private development projects. • Senior Inspector: Associate’s Degree in Civil Engineering Technology or Construction Management and 10 years of progressively responsible experience, of which 5 years are in at a supervisory level.

City of Cocoa Beach Operations Technician C, B or A For more info or to apply

This is technical civil engineering related work for the Water Resources Department performing design, cost estimations and evaluating the construction, alteration or repair of complex public works and capital improvement projects. Work involves preparing preliminary and final engineering designs and specifications through the use of survey data, project drawings and verbal descriptions. Requirements: Bachelor's Degree or a combination of relevant college level education, training and extensive relevant experience; State of FL Class "E" Driver License; considerable knowledge of modern engineering methods and techniques as applied to the design, construction and maintenance of water, wastewater and public works related infrastructure, including performing cost analysis of such projects. Close Date: Open Until Filled; $44,366 - $69,742; See details at EEO-AA-Employer-Vet-DisabledDFWP-Vets' Pref

City of St. Petersburg - Senior Water Resources Manager (Enterprise Services) (IRC#43632) Professional, management work directing the assigned division. Requirements: Extensive experience/knowledge of managing the business, capital improvement programming, and regulatory compliance needs of water and/or wastewater utility programs; Open Until Filled; $87,431$132,911 DOQ; See detailed requirements at EEOAA-Employer-Vet-Disabled-DFWP-Vets' Pref

Florida Water Resources Journal • August 2018


Orange County, Florida is an employer of choice and is perennially recognized on the Orlando Sentinel’s list of the Top 100 Companies for Working Families. Orange County shines as a place to both live and work, with an abundance of world class golf courses, lakes, miles of trails and year-round sunshine - all with the sparkling backdrop of nightly fireworks from world-famous tourist attractions. Make Orange County Your Home for Life. Orange County Utilities is one of the largest utility providers in Florida and has been recognized nationally and locally for outstanding operations, efficiencies, innovations, education programs and customer focus. As one of the largest departments in Orange County Government, we provide water and wastewater services to a population of over 500,000 citizens and 66 million annual guests; operate the largest publicly owned landfill in the state; and manage in excess of a billion dollars of infrastructure assets. Our focus is on excellent quality, customer service, sustainability, and a commitment to employee development. Join us to find more than a job – find a career. We are currently looking for knowledgeable and motivated individuals to join our team, who take great pride in public service, aspire to create a lasting value within their community, and appreciate being immersed in meaningful work. We are currently recruiting actively for the following positions: SCADA Technician, Water Reclamation Division Annual Salary $48,630 Min, $63,242 Mid, $77,854 Max Starting salary of external candidates is customarily below the midpoint based on qualifications. Apply online at: Positions are open until filled.

Town of Davie Assistant Utilities Director $85,823 - $ 99,350 / yr. Utilities Maintenance Supervisor $59,271 - $68,613/ yr. Utilities Electro - Technician $53,552 – $62,104 / yr. Utilities Plant Operator II – Water / Utilities Plant Operator II –Wastewater $47,010 - $54,517/ yr. Please visit our website at for complete job descriptions and to apply. Open until filled.


August 2018 • Florida Water Resources Journal

Utilities, Inc. WATER TREATMENT PLANT OPERATOR Utilities, Inc. of Florida is seeking a Water Operator for the Pasco/Pinellas County area. Applicant must have a minimum Class C FDEP Water license. A dual license is preferred. Applicant must have a HS Diploma or GED & a valid Florida driver’s license with a clean record. To view complete job description & apply for the position please visit our web site,, select the Employment Opportunities tab. Search the Operations & FL, Holiday categories.

City of Groveland Class “C” Water Operator The City of Groveland is hiring a Class "C" Water Operator. Salary Range $ 29,203-43,805 DOQ. Please visit for application and job description. Send completed application to 156 S Lake Ave. Groveland, Fl 34736 attn: Human Resources. Background check and drug screen required. Open until filled EOE, V/P, DFWP

City of Cooper City Utilities Department Utilities Electrician - $49,000 - $69,236 Accepting applications for Utilities Electrician. Performs electrical repairs and preventative maintenance on power and control systems in a water and wastewater utility environment with multiple critical processes. Installs, troubleshoots, and repairs controls and instrumentation. Troubleshoots and repairs electrical systems on heavy industrial equipment up to 4,160 volts. Calibrates and maintains instruments including PH meters, CL2 analyzers, DO meters, and other instruments. Installs new conduit, boxes, panels and pull wire to code as needed to support equipment upgrades. Position open until filled. Use the following link for more information and application: B 4 A- 8 3 5 B 8 C 5 3 A 3 8 F & D E = 3 B 7 6 A 9 8 6 - 2 4 D 3 - 4 3 4 A- 8 B 1 2 CB9C87385B72&Type=B_JOB

Wastewater Treatment Plant Operator Trainee City of Clearwater - Public Utilities Department Manual and semi-skilled work of ordinary difficulty and responsibility in the operation and maintenance of a wastewater treatment plant, sludge handling facility, and grounds on assigned shifts. Trainees work under the supervision of a certified operator, receiving technical instruction in operations, equipment, and maintenance procedures. After becoming familiar with procedures, they may work with independence on certain phases of the process, advancing to Wastewater Plant Operator "C" when proper certification is obtained. See website for description. MINIMUM QUALIFICATIONS: High School graduation, High School Equivalency Diploma, or G.E.D. Certificate. Must have a valid State of Florida Driver's License. Salary: $27,002.84 - $42,140.64 Annually. APPLICATIONS SHOULD BE FILED ONLINE AT:

WATER AND WASTEWATER TREATMENT PLANT OPERATORS U.S. Water Services Corporation is now accepting applications for state certified water and wastewater treatment plant operators. All applicants must hold at least minimum “C” operator’s certificate. Background check and drug screen required. –Apply at or to obtain further information call (866) 753-8292. EOE/m/f/v/d

Utility Compliance/Efficiency Manager $78,836 - $110,929/yr.

Utilities Maintenance Supervisor MAINTENANCE TECHNICIANS U.S. Water Services Corporation is now accepting applications for maintenance technicians in the water and wastewater industry. All applicants must have 1+ years experience in performing mechanical, electrical, and/or pluming abilities and a valid DL. Background check and drug screen required. -Apply at or to obtain further information call (866) 753-8292. EOE/m/f/v/d

$58,829 - $82,778/yr.

Utilities (Safety) Program Coordinator $48,399 - $68,102/yr.

Utilities Storm Water Foreman $47,911 - $67,414/yr.

Utilities System Operator II & III $39,415 - 55,463/yr.; $41,387 - $58,235/yr. Apply Online At: Open until filled.

Generators Enhance Utility’s Hurricane Preparedness In preparation for the upcoming hurricane season, Bonita Springs Utilities Inc. (BSU) has purchased additional tow-behind generators to enhance the utility’s hurricane preparedness capabilities and assist with operations during emergency situations that may result in extended power outages. “The utility weathered Hurricane Irma extremely well last year,” said Bob Bachman, BSU board president. “However, as forecasts call for another busy hurricane season, the board of directors and staff are taking steps to enhance our emergency response capabilities.” The utility’s hurricane planning, action, and infrastructure resulted in successful and sustained operations through Hurricane Irma. Water service was maintained throughout the Bonita Springs and Estero service areas, with one exception. Due to its evacuation zone location and the catastrophic storm surge forecast, BSU made the decision to temporarily cut service to the Bonita Beach area to protect the integrity of the system, private property on the islands, and water supply for the area as a whole. Service was quickly restored, with a standard two-day boil water notice afterward. The preparedness plan for BSU also included switching all plants to generator power before the storm hit in anticipation of losing electricity. All water wells and 332 lift stations throughout the service area lost power. Prior to the storm, permanent generators were installed at a number of these facilities. Afterward, staff worked 12-hour shifts, 24

hours a day, to fuel and pump down the lift stations with a fleet of portable generators until power was fully restored. “Our focus during hurricanes and similar emergencies is on meeting the needs of our community,” said John R. Jenkins, BSU executive director. “Drinking water availability is critical during hurricane recovery activities. Imagine the impact of loss of water and wastewater service, on top of flooding, storm damage, and power outages. We plan yearround in an effort to minimize storm impacts and maximize recovery efforts.” During assessments after Hurricane Irma, BSU identified the need to dedicate more generators to water well and lift station operations during and after a hurricane or other emergency. Three new generators will

be available for identified master lift stations, and a fourth generator will assist in providing backup power to wells in order to maintain potable water service to members. In addition to purchasing generators, BSU continues to pursue ways to harden its facilities and improve its storm preparation. Efforts include: S Expansion of communication capabilities with remote water tanks, wells, and lift stations. S Use of compressed natural gas as an alternative fuel supply for fleet vehicles and equipment. S Sewer system interconnects to afford adequate response time. S A robust sewer system maintenance and replacement program. S

Florida Water Resources Journal • August 2018


January 2016

Editorial Calendar January ..........Wastewater Treatment February..........Water Supply; Alternative Sources March..............Energy Efficiency; Environmental Stewardship April ................Conservation and Reuse; Florida Water Resources Conference May ..................Operations and Utilities Management June ................Biosolids Management and Bioenergy Production July....................Stormwater Management; Emerging Technologies; FWRC Review August ............Disinfection; Water Quality September ......Emerging Issues; Water Resources Management October ..........New Facilities, Expansions, and Upgrades November ......Water Treatment December ......Distribution and Collection Technical articles are usually scheduled several months in advance and are due 60 days before the issue month (for example, January 1 for the March issue). The closing date for display ad and directory card reservations, notices, announcements, upcoming events, and everything else including classified ads, is 30 days before the issue month (for example, September 1 for the October issue). For further information on submittal requirements, guidelines for writers, advertising rates and conditions, and ad dimensions, as well as the most recent notices, announcements, and classified advertisements, go to or call 352-241-6006.

Display Advertiser Index Blue Planet ............................................................................................63 CEU Challenge ........................................................................................13 Florida Aquastore ..................................................................................55 FSAWWA Conference Attendee Registration ........................................24 FSAWWA Conference Exhibit Registration............................................25 FSAWWA Conference Poker/Happy Hour ..............................................26 FSAWWA Conference Golf Tournament ................................................27 FSAWWA Conference Competitions ......................................................28 FSAWWA Conference Water Distribution Awards ................................29 FSAWWA Conference Water Conservation Awards ..............................30 FWPCOA Short School............................................................................43 FWPCOA Training Calendar....................................................................19 FWRC Call for Papers ..............................................................................9 Grundfos ................................................................................................15 Hudson Pump & Equipment ..................................................................31 Hydro International ..................................................................................5 Lakeside Equipment ................................................................................7 Moss Kelley ............................................................................................35 UF Treeo ..................................................................................................39 Stacon ......................................................................................................2 Xylem......................................................................................................64


August 2018 â&#x20AC;˘ Florida Water Resources Journal

Glossary of Common Terms in This Publication ASR ....................aquifer storage and recovery AWT....................advanced water treatment AWWT ..............advanced wastewater treatment AWWA ..............American Water Works Association BOD ..................5-day biochemical oxygen demand BODx..................BOD test based on other than 5 days CBOD ................5-day carbonaceous BOD COD ..................chemical oxygen demand cfm ....................cubic feet per minute cfs ......................cubic feet per second CWA ..................Clean Water Act DEP ....................Fla. Dept. of Environmental Protection EIS......................Environmental Impact Statement EPA ....................U.S. Environmental Protection Agency FAC ....................Florida Administrative Code fps ......................feet per second FSAWWA............Florida Section of AWWA FWEA ................Florida Water Environment Association FWPCOA ..........Florida Water & Pollution Control Operators Association GIS ....................Geographic Information System gpcd ..................gallons per capita per day gpd ....................gallons per day gpm ..................gallons per minute hp ......................horsepower I/I ........................Infiltration/Inflow mgd ..................million gallons per day mg/L ..................milligrams per liter MLSS ................mixed liquor suspended solids MLTSS................mixed liquor total suspended solids NPDES ..............Nat. Pollutant Discharge Elimination System NTU ....................nephelometric turbidity units ORP....................oxidation reduction potential POTW ................public-owned treatment works ppm per million ppb per billion PSC ....................Public Service Commission psi ......................pounds per square inch PVC ....................polyvinyl chloride RO ......................reverse osmosis SCADA................supervisory control and data acquisition SJRWMD............St. Johns River Water Mangement Dist. SFWMD ..............South Florida Water Management Dist. SRWMD..............Suwannee River Water Management District SSO ....................sanitary sewer overflow SWFWMD ..........Southwest Fla. Water Management Dist. TDS dissolved solids maximum daily load TOC organic carbon TSS suspended solids USGS ................United States Geological Survey WEF....................Water Environment Federation WRF ..................water reclamation facility WTP....................water treatment plant WWTP ................wastewater treatment plant

Florida Water Resources Journal - August 2018  

Disinfection and Water Quality

Florida Water Resources Journal - August 2018  

Disinfection and Water Quality