Renewable Energy for the Expanded Joe Mullins

Water Treatment Plant in Melbourne, Florida

Katie Welsh, Alana Stevens, Alexis Mendez,

Brice Peters, Mateo Arimany, Cameron Roberts, Shadrach Elechi, Rebecca Kitto, Siri Swangjai,

William Lee, Paige Wachtler, Takashi Kida, Samantha Perry, Katie Burke, Casselle Russell,

and Juan Pablo Aljure

Florida Institute of Technology Department of Marine and Environmental Science

ABSTRACT The Joe Mullins Reverse Osmosis (R.O.) Plant currently

supplies a customer base of 135,000 consumers in Melbourne,

Florida with water. It is projected that the consumer base will rise

to 208,000 by 2025 (2000/2007 Census Bureau). The current

R.O. plant has the capacity to supply the 15.7 million gallons per

day that is consumed by the customer base in 2009, but needs to

be expanded in order to ensure that the predicted 24.1 million

gallons per day in 2025 can be met. The expansion will also

allow for additional water from Lake Washington to be utilized by

the northern counties that are more populated due to tourism. The

renewable energy would provide one third of the required energy

with a reasonable water cost per 1000 gallons.

INTRODUCTION

The Joe Mullins Reverse Osmosis (R.O.) Plant currently

supplies the Melbourne, Florida customer base 6.5 MGD of water

with 3 artisan wells (2 MGD each) and has a 1.5 MGD brine

discharge maximum. It is proposed that the plant be expanded to

supply 16 MGD with 10 artisan wells with a 4.0 MGD brine

discharge maximum. The St. John’s River Water Management

District proposes that the expansion would allow northern

counties in Florida to consume additional water, and lower

Melbourne’s consumptive use to 8 MGD or less. To improve the

current R.O. plant to reach the energy capacity needed to produce

the proposed 16 MGD, renewable energy sources in the form of

PV modules and batteries will be used to pump the wells and

brine discharge, operate the plant, and include an energy backup

system. Two different options will be explored in a manner where

33% of the power will come from the PV modules and/or

batteries, and the remaining power will be acquire from Florida

Power and Light Company (FPL). The required permits from the

Florida Department of Environmental Protection, St. Johns River

Water Management District and other relevant authorities will

also be explored along with the environmental impacts of the

renewable energy design and discharge of the brine. A proposed

budget including the possibility of funding and a comparison to

current water usage prices and operation of similar plants will also

be included to determine the overall feasibility of the expansion of

the R.O. plant in a term of twenty years.

TECHNICAL SPECIFICATIONS

The technical specifications are designed to produce

33% of the required energy, and will explore the use of two

different options. The remaining energy that is not provided by

renewable sources will be acquired through FPL.

Option 1 presents a design that will use only PV

modules to produce all of the required energy to operate the plant,

pump the wells, and discharge the brine, and option 2 uses a

combination of PV modules and batteries to increase the overall

function of the PV modules by increasing the hours available from

the sun from 5.67 hours to 8.5 hours (See tables1 and 2 below).

Both options use the same basic design of a square area that

allows 21 ft² per panel with 4.9 ft. between rows to ensure that the

panels do not shadow one another and lessen the total energy

production. Both options also require that the PV panels face

south and are tilted at 28º to receive maximum sun hours year

round (See basic technical design figures 1 and 2).

In comparison, both options will be able to produce the needed energy to operate the plant, and are also feasible in an economical stand point; however, option 2, which uses batteries, will require less acres of land for the set up, and also creates a profit for the R.O. plant rather than a deficit. Table 1: Energy and area required for consumption without batteries.

Figure 1: Technical design of the space needed for each PV module, and the area needed to reduce shadow overlap.

Figure 2: Technical design of the angles and design needed to obtain the optimum amount of sun hours from the PV panels. Table 2: Energy and area required for consumption with batteries.

Circuit Design (no batteries)

Option 1

Unit energy required 11 KWh/KGal

Maximum daily water processed 16,000 Kgal

Maximum daily energy required 176,000 KWh Energy needed by the plant from the PV array (33%) 58,667 KWh Energy supplied by PV Array before inverter with 97% eff. 60,481 KWh

Sun hours with no batteries available 5.67 hours

Power supplied from PV array 10,667 KW

280 W Panels needed 38,096 Area needed for PV array @ 21 ft2/panel + separations 22 acres

Circuit Design (with batteries) Option 2

Maximum daily water processed 16,000 Kgal

Maximum daily energy required 176,000 KWh Energy needed by the plant from the PV array (33%) 58,667 KWh Energy required by batteries before inverter with 97% eff. 60,481 KWh Hours availability in total from sun-hours and batteries 8.5 hours min

Power required after photocontroller 7,115 KW Power needed by PV array before 97% eff. Photocontroller 7,333 KW

280 W Panels needed 26,191 26,191 Area needed for PV array @ 21 ft2/panel + separations 16 acres

BUDGET AND FUNDING

Budgeting and funding for each option is outlined in a

plan that would include financing the entire project for a term of

twenty years in order to make the overall cost more economical

and affordable for the city of Melbourne.

The total cost for option 1, which does not include the

batteries, is $358.8 million (See figure 3 for itemized cost

breakdown). The total average of water that is predicted to be

produce in twenty years is 93,440 million gallons, which would

suggest that the total cost to consume water would be $3.84 per

K/gal (See figure 4). In comparison to the current cost of $3.67

per K/gal, the cost would be greater if produced by the PV panels

on in option 1 and would leave the city with a $0.17 per K/gal or

$794,240.00 deficit that would need to be offset with grants or an

alternate type of funding (See figure 5).

The total cost for option 2, which includes batteries in

the design, is $332.7 million (See figure 6 for itemized cost

breakdown). The total average of water that is be predicted to be

consumed in twenty years is 93.440 million gallons, which would

suggest that the total cost to consume water would be $3.56 per

K/gal (See figure 7). In comparison with the current cost of $3.67

per K/gal, the cost is less and would not require any grants or

alternate funding to complete the project, and will allow the R.O.

plant to profit instead of face a deficit (See figure 8).

Figure 3: Itemized cost over twenty years without batteries.

Figure 4: Itemized cost per K/gal of water produced without batteries.

Figure 5: Comparison cost of current water consumption prices with predicted prices of the expanded R.O. plant without batteries.

Figure 6: Itemized cost over twenty years with batteries.

Figure 7: Itemized cost per K/gal of water produced with batteries.

Figure 8: Comparison cost of current water consumption prices with predicted prices of the expanded R.O. plant with batteries.

ENVIRONMENTAL IMPACT

Another key concern regarding the expansion of the Joe

Mullins Reverse Osmosis Water Treatment Plant is the impacts it

would have on the environment regarding both solar array

construction on the land, and the disposal of an additional 2.5

million gallons of brine due to the expansion. When considering a

model of solar panel for a renewable energy to fuel the plant two

mainstream prototypes are recommended: the cost-effective

cadmium panels, and the more reliable Poly-Crystalline solar

panels.

One major benefit of the cadmium-based PV array is the

sheer amount of electricity produced by the set-up. There are

about 20,000 MT of Cd used each year, half of which, is used for

Ni/Cd batteries. If used for PV panels, 20,000 MT/yr could

supply 1010 m2 of new PV annually. At 10% sunlight-to-

electricity efficiency, 1,000 GWp of new PV capacity per year

would be produced and approximately 1.6 million GWh of energy

a year. The U.S. utility grid produces around 3 million GWh a

year so it would only take two years of using the world’s Cd to

replace the entire U.S. grid.

Although the cadmium based arrays are cheaper, the

hazardous material used to manufacture them minimizes their

potential use considerably. The cadmium that is used in

producing the arrays is a byproduct of copper, lead, and zinc

mining which can be extremely harmful for both humans and the

ecosystem. Cadmium also poses a threat of contaminating Lake

Washington, the source of water for the treatment plant. It is also

a toxic material that can lead to kidney and breathing problems

according to the U.S. Labor Department Doctors recommend

seeking treatment if blood levels of a human reach levels of

cadmium over 5 µg/l in the blood. Amounts used in these arrays

include:

Amount of Cadmium in CdTe Layers of Various Thickness

Microns 5 2 1 0.5 0.2

g/m² 14 g 5.5 g 2.75 g 1.4 g 0.55 g

This table illustrates the possibility of a PV repairman c

accumulating cadmium poisoning from long-term exposure with

the substance in the arrays.

Another common substance found in cadmium based

solar panels is arsenic. Arsenic can easily contaminate

groundwater used for drinking. It could also provide an

occupational hazard to personnel hired to examine the PV

modules. The physical and environmental impacts that occur

from cadmium panel use outweigh the cost-effective benefits.

Poly-Crystalline solar arrays, although pricier, are much

more reliable and environment friendly. The silicon material they

are made of provides greater strength and require less repair.

Although, the PVs are safe while in use, the extraction of

materials used to manufacture them may cause environmental

disturbances and occupational hazards due to the mining of silica

particles. If inhaled, these particles have the potential to cause

lung disease. In addition, the production of the silicone panels

require the use of fluorine, chlorine, nitrates, sulfur dioxide,

nitrogen oxide, carbon dioxide, and more silica particles that

could be dangerous for employees.

While in use, the Poly-Crystalline solar panels have the

least impact on the environment including water, land, and sea,

but natural disasters such as summer hurricanes and fires, which

are common to Brevard County, must be taken into consideration.

Ingress of water into the panel is not a problem because these PV

models are sealed to prevent the corrosion of metal contacts which

could cause a system failure. In most cases, because, they can be

successfully sealed, the panels have a 20 year warranty.. This

highly effective sealing also prevents the module from leaching

material to the environment. These systems are so stable, that it is

almost impossible to break them into many smaller pieces. In

fact, CdTe and other semiconductors are some of the most stable

and impervious materials found. This stability also reduces the

risk of fire damage. Unless, trace elements are found on the

outside of the model due to accidental spills during the

manufacturing process, fires are not an issue.

A second, and just as important environmental concern

is the disposal of the additional brine waste water. After

expansion, brine quantities will increase from 1.5 million gallons

a day to 4.0 million gallons a day. Three possible solutions that

will be explored for discarding the waste are deep-well injection,

off-shore dumping into the Atlantic Ocean, and creating an

evaporation pond.

Because of ecosystem destruction, it is not likely that

the Melbourne Water Management company will allow for an

increase in brine disposal into the Eau Gallie River. The possible

impacts for discharging addition brine include a possible

degradation in water quality, a change in the designated use

regarding the Eau Gallie River, and a negative alteration in the

habitat of fish, crustaceans, sea grasses, and sediments.

Before deep-well injection can occur certain

requirements are necessary before permits will be issued. First,

the site for the future injection must be thoroughly examined to

determine whether it is geologically suitable. It must also be

inspected to ensure that the brine waste does not migrate into a

source of underground drinking water. A plan must be set for

closing the well and financial assurance must be observed. Lastly,

the site must be ensured that the pressure caused by the injection

process itself does not cause fractures in the injection zone or the

migration of fluids for the next 10,000 years.

There are some hazards pertaining to the process of

deep-well injection that must be addressed. The land that

surrounds the Joe Mullins Reverse Osmosis Water Treatment

Plant is on Florida Karst Topography. This means that the

geography is limestone based with many cracks and fissures that

may lead to a sinkhole formation. There is also the possibility of

a leak from the seepage of the brine from its confining layer

through the cracks and fissures into the superficial aquifer. In

such a situation, the brine would increase the salinity and add

other contaminants into the water and sediment.

Some environmental and man-made solutions have

facilitated the process. For example, geographically, the land is

separated into many poorly permeable layers that prevent the

brine and drinking waters to mix. Three or more protective layers

of pipe or tubing around the well shaft that goes into the injection

site ensure that the system will not leak into the surrounding

ground layers. Lastly, protective layers between the well and the

drinking aquifer will be established to also prevent leakage.

Below is a diagram of the deep-well injection set-up:

Figure 9: Deep-well injection set-up.

The alternative for deep-well injection for the disposal

of the brine waste water is to dump it offshore in the Atlantic

Ocean. Unfortunately, impacts on the marine ecology make this

scenario less appealing. Not only are salinity gradient changes

and chemical additives a possible problems, but also the weight of

the brine can cause it to sink rather than dissolve resulting in a salt

scar. A salt scar limits nutrient development and thus, life

development. Equipment corrosion can also occur resulting in the

release of titanium, copper, and nickel into the sea. These

substances directly harm marine organisms and humans

inadvertently due to consumption of contaminated fish and

shellfish. The most important environmental concern with

offshore-dumping is the impact on a potentially sensitive

ecosystem that would quickly diminish around the site of disposal.

The first to be affected would be the marine biota such as

phytoplankton and algae that provide the basis for organisms in

the ocean.

A third possible alternative would be to create an

evaporation pond. This concept involves placing the brine waste

water into a shallow pool of water and letting the sun evaporate it

until a salt substance remains that can be used as nutrients or for

other means. A problem with this alternative is the rainfall on

Florida’s east coast would make the process almost impossible;

therefore, deep well injection would be the most reasonable and

environmentally friendly method of disposing the waste.

PERMITS AND REGULATIONS

The Joe Mullins Reverse Osmosis Water Treatment

Plant must adhere not only to the standards and regulations set by

the Environmental Protection Agency (EPA), but must also obtain

the appropriate permits issued by the National Pollutant Discharge

Elimination Systems (NPDES), the Florida Department of

Environmental Protection (FDEP), and the St. Johns River Water

Management District in order to run its facility.

One permit that is necessary for water removal,

treatment, and exportation to consumers is the Consumptive Use

Permit (CUP). The St. Johns River Water Management District

requires all companies that utilize large quantities of water to

obtain this permit. Currently, Florida is divided into five

management districts. The St. Johns Water Management District

has jurisdiction over Brevard County including the Indian River

Lagoon, Eau Gallie River, and Lake Washington. The city of

Melbourne had been operating under CUP permit #50301 issued

on May 11, 1999 until recent concerns regarding brine disposal

from the reverse osmosis process was addressed. Since 2006,

however, the plant was issued a Temporary Consumptive Use

Permit (TCUP). Beginning that year, the processing plant was

required to filter greater amount than the permit allotment of

surface water from Lake Washington to meet the public supply

needs. The TCUP allows the plant to use up to 15.5 million

gallons per day of withdrawals and although the permit expired

October 11, 2006, Melbourne has continued to use this amount.

The expansion of the Joe Mullins Reverse Osmosis

Water Treatment Plant from three wells to ten requires additional

permits including the environmental resource permit (ERP). The

environmental resource permitting program is designed to ensure

that new construction will not adversely affect water flow or

storage, thus, preventing flooding. ERP combines two permits:

the former Wetland and Dredge Fill permit issued by the Florida

Department of Environmental Protection FDEP) and the

Management and Storage of Surface Waters permit issued by the

water management districts. An ERP is required by anyone

proposing construction including government agencies.

Once an ERP is obtained the Florida Department of

Environmental Protection must be contacted in order to obtain the

appropriate permits required for water well construction and for

the delegation of a contractor-licensing program in the St. Johns

River Water Management District. The District established these

construction standards to ensure that newly constructed water

wells do not cause uncontrolled water flow or water quality

degradation. The District also issues water well construction

permits to FDEP-delineated groundwater contamination areas.

Further permits are required for the operation of the Joe

Mullins Reverse Osmosis Water Treatment Plant. These permits

are applied for and granted under FDEP 62-4 permits: Drinking

Water and Ground Water Rules. Within these rules there are

further regulations that must be adhered to when operating a water

treatment plant including well dimensions, staffing, and brine

waste water disposal. The following permits would be necessary:

Under the FDEP Drinking Water Regulations

• 62-550: Drinking Water Standards, Monitoring and

Reporting

• 62-602: Drinking Water and Domestic Wastewater

Treatment Plant Operators

• 62-699: Treatment Plant Classification and Staffing

FDEP Ground Water Regulations

• 62-520: Ground Water Classes, Standards, and

Exemption

• 62-522: Ground Water Permitting and Monitoring

Requirements

• 62-528: Underground Injections Control

• 62-521: Wellhead Protection

In addition to permits required for the removal and

purification of the surface waters, construction permits required

by the FDEP include:

• 62-555: Permitting and Construction of Public Water

Systems

• 62-560: Requirements for Public Water Systems that are

Out of Compliance

• 62-531: Water Well Contractors

• 62-532: Water Well Permitting and Construction

Requirements

Regarding the disposal of brine waste water, The Joe

Mullins Reverse Osmosis Water Treatment Plant holds a shared

permit. This shared permit includes an Underground Injection

Control Permit from the FDEP along with a discharge permit

through the National Pollutant Discharge Elimination Systems.

The latter is authorized by the Clean Water Act under the EPA

and is responsible for the control of water pollution by regulating

point sources that discharge pollutants into the waters of the

United States. Currently, the water treatment plant is permitted to

dispose of 1.5 million gallons of the brine waste a day into the

Eau Gallie River.

In addition, the EPA prescribes regulations that limit the

amount of contaminants into the water provided by public water

systems. It further requires annual reports covering water quality

from the facility. These water quality tests must be performed by

a state-certified laboratory that continuously analyzes quality

throughout the entire treatment process to ensure the population of

no pollutants. These annual reports can be viewed at:

www.melbourneflorida.org. Also through the EPA, the

Underground Injection Control (UIC) program is responsible for

the regulation of all construction, operations, permitting, and

closure of injection wells that place fluids underground for

disposal or storage. Under the UIC, further permits are necessary

for expanding the existing water treatment plant.

The FDEP also performs water quality assessments

through another program: Source Water Assessment and

Protection Program (S.W.A.P.P.). This program ensures that the

drinking water is safe at its source. The Florida Department of

Environmental Protection is initiating the S.W.A.P.P. as part of

the federal Safe Drinking Water Act (SDWA). Results of this

assessment are located at the Florida Department of

Environmental Protection website:

http://www.dep.state.fl.us/swapp/.

CONCLUSION

Providing 33% of the energy needed to run the

expanded Joe Mullins Reverse Osmosis Water Treatment Plant is

technically feasible and economically viable. Of the two potential

solar power setups, option 2, which integrates batteries into the

design, is not only less expensive over the next 20 years but, it

requires no grants for funding and is able to produce the same

amount of power. Option 2 also utilizes less acreage, fewer solar

panels, and increased availability to utilize the maximum amount

of renewable energy. This is only feasible if the battery setup is

economically viable within the unit cost shown.

By installing 26,191 arrays over 16 acres and using

batteries, the company will essentially be profiting from the

expansion in a long term sense. The total cost for option 2 over

the next 20 years is $332.7 million dollars. This statistic

encompasses capital costs, costs to Florida Power and Light for

providing the additional 66% of the energy, the price and

installments of the PVs, and the maintenance required to run the

facility. By producing the drinking water for $3.56 per Kgal and

selling it for the standard $3.67 per Kgal, the company would

profit $0.11 per Kgal of water consumed.

After discussing various possibilities for disposing of an

extra 2.5 million gallons a day of brine waste water, the best

alternative suggested is deep well injection. Certain safety

precautions must be considered before initializing the deep well

injections to prevent possible sinkholes or brine backwash into

drinking aquifers. A poorly permeable layer must be placed

between the two and a soundly structured well must be

constructed.

To expand the R.O. plant from three wells to ten wells,

certain permits and regulations must be upheld to ensure safety,

quality, and habitat maintenance. The St. Johns River Water

Management District requires an Environmental Resource Permit,

certain water well permits, and a licensed contractor. The

Environmental Protection Agency must provide an Underground

Injection Control permit for proper brine disposal. Florida

Department of Environmental Protection requires over 11 permits

for optimum insurance over water quality, maintenance safety,

well soundness, and construction compliances related to both

drinking water and groundwater permits.

For future studies, perhaps a water turbine can be placed

in the reverse osmosis facility to create energy from the current

produced by the brine waste water. In upcoming years, the

demand by consumers for drinking water in Brevard County will

increase as Lake Washington slowly decreases. This expansion

must be viewed as only a temporary fix for the economic

problems at hand in the near future. We suggest a more reliable

source of drinking water supply by introducing a desalination

plant at a Melboure/Palm Bay site, which would take water from

the Indian River Lagoon and discharge it into the Atlantic Ocean.

Although this would be a billion dollar investment, it would

provide climate ready facilities, appropriate amounts of drinkable

water for many more years, and for a $3.17 K/gal unit cost

(Lindler and Aljure, 2009). required 11 KWh/K

ACKNOWLEDGMENTS

Our thanks to Mr. Frank Leslie for direction and

assistance in the overall outcome of the project. Mr. David

Phares, Assistant Superintendent, and Mr. Cody Wells,

Operations Supervisor, of Water Production for the City of

Melbourne for providing relevant data and information about the

current plants, and possible future scenarios for future solutions.

Ms. Colleen Lindler, Florida Institute of Technology graduate

student, for providing environmental issues information. Ms.

Krista Simon, Tampa Bay Water Records Manager, for providing

two CDs with information about the Tampa Bay desalination

plant.

REFERENCES

1. Alsema, E.A., de Wild-Scholton, M.J., & Fthenakis, V.M. (2006). Environmental Impacts of PV Electricity Generation: A critical comparison of energy supplyoptions. Available at http://www.clca.columbia.edu/papers/21-EUPVSC-Alsema- DeWild-Fthenakis.pdf

2. Bowman, M., Debray, S. K., and Peterson, L. L. 1993.

Reasoning about naming systems. ACM Trans. Program. Lang. Syst. 15, 5 (Nov. 1993), 795-825. DOI= http://doi.acm.org/10.1145/161468.161471.

3. Brown, L. D., Hua, H., and Gao, C. 2003. A widget framework for augmented interaction in SCAPE. In Proceedings of the 16th Annual ACM Symposium on User interface Software and Technology (Vancouver, Canada, November 02 - 05, 2003). UIST '03. ACM Press, New York, NY, 1-10. DOI= http://doi.acm.org/10.1145/964696.964697

4. Ding, W. and Marchionini, G. 1997 A Study on Video

Browsing Strategies. Technical Report. University of Maryland at College Park.

5. Dowe, Mark. “2007 Annual Drinking Water Quality

Report.” Barefoot Bay Water Treatment System. Brevard County Utility Services, 2007. http://www.brevardcounty.us/usd/geninfo.cfm

6. Florida Department of Environmental Protection. “Source Water Assessment & Protection Program.” June 15, 2007. http://www.dep.state.fl.us/swapp/

7. Forman, G. 2003. An extensive empirical study of feature selection metrics for text classification. J. Mach. Learn. Res. 3 (Mar. 2003), 1289-1305.

8. Fröhlich, B. and Plate, J. 2000. The cubic mouse: a new device for three-dimensional input. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (The Hague, The Netherlands, April 01 - 06, 2000). CHI '00. ACM Press, New York, NY, 526-531. DOI= http://doi.acm.org/10.1145/332040.332491

9. Fthenakis, V.M., Kim, H.C., & Alsema, E.A. (2008). Emissions from Photovoltaic Life Cycles.

Environmental Science and Technology. 42, 2168 – 2174. Available at http://pubs.acs.org/cgi-bin/abstract.cgi/esthag/2008/42/i06/abs/es071763q.html

10. National Pollutant Discharge Elimination Systems

(NPDES). “NPDES Permit Program Basics.” Environmental Protection Agency. December 15, 2008. http://cfpub.epa.gov/npdes/home.cfm?program_id=45

11. Public Works & Utilities Water Supply. “City of Melbourne’s 2007 Annual Drinking Water Quality Report.” City of Melbourne. 2007. http://www.melbourneflorida.org/watercon/water.htm

12. Sannella, M. J. 1994 Constraint Satisfaction and Debugging for Interactive User Interfaces. Doctoral Thesis. UMI Order Number: UMI Order No. GAX95-09398., University of Washington.

13. Spector, A. Z. 1989. Achieving application requirements. In Distributed Systems, S. Mullender, Ed. Acm Press Frontier Series. ACM Press, New York, NY, 19-33. DOI= http://doi.acm.org/10.1145/90417.90738

14. St. Johns River Water Management District. “Water resource permitting information.” 2009. http://sjr.state.fl.us/regulatory/permitinformation.html

15. Svennson, Marcus. “Desalination and the Environment:

Options and considerations for brine disposal in inland and coastal locations.” SLU. Aqualyng: September, 2005.

http://www.jpaljure.com/images/stories/FIT/Environmental_Team_brine_disposal

16. Y.T. Yu, M.F. Lau, "A comparison of MC/DC, MUMCUT and several other coverage criteria for logical decisions", Journal of Systems and Software, 2005, in press.