<%BANNER%>

Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-05-31.

Permanent Link: http://ufdc.ufl.edu/UFE0024111/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-05-31.
Physical Description: Book
Language: english
Creator: Munksgaard, Kim
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, M.E.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Statement of Responsibility: by Kim Munksgaard.
Thesis: Thesis (M.E.)--University of Florida, 2009.
Local: Adviser: Sansalone, John.
Electronic Access: INACCESSIBLE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024111:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024111/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-05-31.
Physical Description: Book
Language: english
Creator: Munksgaard, Kim
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, M.E.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Statement of Responsibility: by Kim Munksgaard.
Thesis: Thesis (M.E.)--University of Florida, 2009.
Local: Adviser: Sansalone, John.
Electronic Access: INACCESSIBLE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024111:00001


This item has the following downloads:


Full Text

PAGE 1

1 PARTITIONING AND TRANSPORT OF POLLUTANTS IN RAINFALL-RUNOFF FROM INDUSTRIAL AND PARKING LOT SOURCE AREA WATERSHEDS By KIM MUNKSGAARD A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Kim Munksgaard

PAGE 3

3 To my Dad who has always supporte d me in my pursuit of education

PAGE 4

4 ACKNOWLEDGMENTS I have m uch gratitude, thanks, and appreciation that I would like to give to those people who helped me successfully complete my research. I would like to thank first and foremost Dr. Sansalone, my advisor, for al ways providing the data, tools, advice, and encouragement I needed. Additionally I would like to give my appreciation to my comm ittee who also provided feedback on my progress of my publications and ma sters thesis. I would like to thank Dr. Kim for giving me insight on topics where I needed fu rther clarification and po inting me in the right direction for resources. Finally, I would like to thank my family and friends who have always given me support emotionally in my pursuit of education.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................7LIST OF FIGURES .........................................................................................................................8LIST OF ABBREVIATIONS ........................................................................................................ 10ABSTRACT ...................................................................................................................... .............12CHAPTER 1 INTRODUCTION .................................................................................................................. 14Concentrations of Pollutants According to Land Use ........................................................... 14Biogenic and Anthropogenic Sources ................................................................................... 15Partitioning of Pollutants .................................................................................................... ... 16First Flush Phenomenon ........................................................................................................ 16Health Effects of Pollutants ................................................................................................... 18Bioavailability of Metals ....................................................................................................... 19Consequences of Nutrient Loading ....................................................................................... 20Implications of Stormwater Characteristics ........................................................................... 202 PARTITIONING AND TRANSPORT OF METALS IN RAINFALL-RUNOFF FROM AN INDUSTRIAL SOURCE AREA WATERSHED ........................................................... 21Introduction .................................................................................................................. .......... 21Objectives .................................................................................................................... .......... 23Background .................................................................................................................... ........ 23Site Description ..................................................................................................................... 26Methodology ................................................................................................................... ....... 27Rainfall-Runoff Measurements and Sample Collection ..................................................27Analytical Procedures ......................................................................................................28Results .................................................................................................................................... 29Hydrology ..................................................................................................................... ...29M(V) Curves: Metals .......................................................................................................30M(V) Curves: TDS and TSS ...........................................................................................31M(V) Curves: Other Pollutants .......................................................................................32EMC ........................................................................................................................... .....33Fd and Kd ..........................................................................................................................34Implications/Discussion ....................................................................................................... .. 35Conclusions ............................................................................................................................ 36

PAGE 6

6 3 NUTRIENT PARTITIONING AND TRANSPO RT IN RAI NFALL-RUNOFF FROM AN INSTITUTIONAL SOURCE AREA .............................................................................. 56Introduction .................................................................................................................. .......... 56Objectives .................................................................................................................... .......... 57Background .................................................................................................................... ........ 58Saint Johns River Water Management District (SJRWMD) ..........................................58Nutrient Concentrations from Paved Surfaces ................................................................ 59Mass-Limited vs. Flow-Limited Delivery ....................................................................... 59Site Descriptions .................................................................................................................... 60Methodology ................................................................................................................... ....... 61Stormwater Sample Collection ........................................................................................61Traffic Count ...................................................................................................................61Nutrient Partitioning ........................................................................................................62EMC Calculations ...........................................................................................................63Mass-Limited vs. Flow-Limited Behavior ...................................................................... 63Particle Size Distribution (PSD) ......................................................................................64Results .................................................................................................................................... 64Hydrology ..................................................................................................................... ...64Traffic Count ...................................................................................................................65M(V) Curves: Particulate Fracti ons and Particulate N and P ..........................................65M(V) Curves: TDS and Dissolved N and P ....................................................................65Fractions of PM ...............................................................................................................66Nutrient Partitioning ........................................................................................................66Total Site Nutrient Loads ................................................................................................67EMC of Water Chemistry Parameters ............................................................................. 67Implications/Discussion ....................................................................................................... .. 68Conclusions ............................................................................................................................ 684 CONCLUSIONS ................................................................................................................... .88Pollutant Concentrations ........................................................................................................ 88Pollutant Loads ...................................................................................................................... 88Partitioning of Pollutants .................................................................................................... ... 88Transport of Pollutants .......................................................................................................... 89Industrial Water Reuse .......................................................................................................... 89Irrigation Reuse Water ........................................................................................................ ... 92LIST OF REFERENCES ...............................................................................................................97BIOGRAPHICAL SKETCH .......................................................................................................104

PAGE 7

7 LIST OF TABLES Table page 2-1 Hydrologic indices and pavem e nt residence time data ...................................................... 492-2 Cumulative mass for metals ............................................................................................... 502-3 Values of k0 and k1 for Cu, Pb, Zn, and Ag for both catchments ..................................... 512-4 Cumulative mass fraction maximum values fo r various water chemistry parameters ...... 522-5 Values of k0 and k1 for TDS and TSS for both catchments ............................................... 532-6 Summary of particulate and dissolved metal EMCs .......................................................... 542-7 Summary of EMCs for runoff chemistry parameters ........................................................553-1 Summary of hydrology ......................................................................................................833-2 Cumulative total of phosphorous, nitrogen, and particulate matter ...................................843-3 Values of k0 and k1 for particulate-bound N and P, and PM ............................................853-4 Values of k0 and k1 for PO43-, NO3-, and TDS ...............................................................863-5 EMC of particulate and dissolved pho sphorous, nitrogen, and particulate matter ............ 874-1 Water quality requirements for use in steam generation and cooling in heat exchangers. ................................................................................................................... ......954-2 Recommended maximum concentrations of water quality parameters in irrigation water ......................................................................................................................... ..........96

PAGE 8

8 LIST OF FIGURES Figure page 2-1 Plan view of watershed located in Great er Los Angeles with delineation of W MIC and EMIC and drainage flow paths ................................................................................... 382-2 Hydrology hyetographs, cumulative and incr emental hydrographs as a function of time for the storm events. Time 0 represents start of rainfall. .......................................... 392-3 Runoff coefficients for storm ev ents for both WMIC and EMIC ...................................... 402-4 Plots of cumulative mass fractions as a function of cumulative runoff volume for metals ........................................................................................................................ .........412-5 Plots of cumulative mass fractions as a function of cumulative runoff volume for TDS and TSS .....................................................................................................................422-6 Plots of cumulative mass fractions as a function of cumulative runoff volume for DOC, TOC, ALK as CaCO3, and Hardness as CaCO3 ......................................................432-7 Conductivity, pH, turbidity, and redox incremental plots as a function of time for the 20 December 2002 and 12 February 2003 storm events .................................................... 442-8 Conductivity, pH, turbidity, and redox incremental plots as a function of time for the 24 February 2003 and 15 March 2003 storm Events ......................................................... 452-9 Comparison of the sample concentrations of Cu, Zn, Pb, and Cd, as well as TSS at the greater Los Angeles site to other urban and industrial sites Data Not Available ..... 462-10 Dissolved fraction and partitioning coefficient statistics for metals .................................. 472-11 Comparison of the sample fd and kd of Cu, Zn, Pb, and Cd at the Greater Los Angeles site to other urban and sites ..................................................................................483-1 Plan view of catchment .................................................................................................... ..713-2 Event-based hydrology ..................................................................................................... .723-3 Event-based rainfall-runoff coefficient ..............................................................................733-4 Traffic count for catchment basin during a typical wee kday and weekend. ......................743-5 Mass-limited behavior of particulat e fractions, and part iculate N and P ........................... 753-6 Mass-limited and flow-limited behavior of dissolved Nitrogen, Phosphorous, and Total Dissolved Solids .......................................................................................................76

PAGE 9

9 3-7 Fraction of suspended, settleable, and se dim ent vs. the maximum SSC concentration throughout the hydrograph ................................................................................................. 773-8 Event-based particle si ze distribution of PM ..................................................................... 783-9 Dissolved fraction and equilibrium pa rtitioning coefficient for Phosphorous and Nitrogen .............................................................................................................................793-10 Comparison of dissolved fraction and partitioning coefficient for other similar sites that represent parking of pavement .................................................................................... 803-11 Total annual nutrient loads of catchment basin for total phosphorous, nitrogen, and PM based on the high and low runoff coefficients ............................................................813-12 Comparison of particulate and dissolved nitrogen and phosphorous to other similar sites that represent parking of pavement ............................................................................ 82

PAGE 10

10 LIST OF ABBREVIATIONS Scale factor BMP Best management practices c(t) Time variable dissolved or particulate-bound concentration (mg/L) C Runoff coefficient COC Constituents of concern CTR California toxics rule DO Dissolved oxygen DOC Dissolved organic carbon (mg/L) EMC Event mean concentration EMV Event mean value EPA Environmental Protection Agency fd Dissolved fraction FDEP Florida Department of Environmental Protection HS Hydrodynamic separator k0 Zero-order coefficient k1 First-order coefficient kd Partitioning coefficient (L/kg) K Transport rate coefficient M TSS concentration of the adsorbing solid material in the aqueous system (mg/L) Mt Cumulative mass delivered (mg) M0 Constituent mass on the surface at th e beginning of the rainfall-runoff event (mg) M Total mass of constituent over entire event duration (mg)

PAGE 11

11 M(V) Cumulative pollutant mass as a f unction of cumulative discharged volume curve (mg) N Nitrogen (mg/L) ND Not detectable NO3 Nitrate (mg/L) P Particulate-bound mass of a heavy metal (mg) P Phosphorus (mg/L) PM Particulate matter PSD Particle size distribution q(t) Time variable flow (L/s) Q50 Median flow rate (L/min) Qmax Maximum flow rate (L/min) SJRWMD Saint Johns River Water Management District SSC Suspended solids concentration (mg/L) tc Initial lag time (min) tmean Mean lag time (min) TDS Total dissolved solids (mg/L) TMDL Total maximum daily load TN Total nitrogen (mg/L) TOC Total organic carbon (mg/L) TP Total phosphorus (mg/L) TSS Total suspended solids (mg/L) Vt Cumulative volume (L) V Total volume of flow over entire event duration (L) WMIC West major interior collector WQO Water quality objectives

PAGE 12

12 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering PARTITIONING AND TRANSPORT OF POLLUTANTS IN RAINFALL-RUNOFF FROM INDUSTRIAL AND INSTITUTIONAL SOURCE AREA WATERSHEDS By Kim Munksgaard May 2009 Chair: John J. Sansalone Major: Environmental Engineering Sciences Anthropogenic activities and interfaces superimposed on the rainfall-runoff process result in the generation and transport of chemical and particulate loads. This research focused first on partitioning and transport of metals as well as particulate matter (PM) from a 73 ha industrial source area watershed located in Los Angeles County, California. Specific constituents of interest in runoff included phase (dissolved and particulate) fraction of Cu, Ag, Pb, and Zn, PM fractions, and the water chemistry indices. Trans port classifications, event-mean concentrations (EMCs), dissolved fractions, and partitioning coefficients were ex amined. Results indicate that particulate-bound metals and PM (as total suspended solids, TSS) exhibited mass-limited transport with a median k1 value of 1.802, while the dissolved metal fractions and total dissolved solids (TDS) demonstrated flow-lim ited behavior, with a median k0 value of 0.979. Phase fractionation indicated that Pb was primarily bound to PM, with a mean dissolved fraction value of 0.036, while Cu and Zn partitioned more equally between phases. The site mean concentrations for Cu, Pb, Ag and Zn were 55.9, 181.4, 1.3, and 143.7 micrograms per liter respectively, while EMCs were higher than California Toxics Rule (CTR) and were comparable to levels found in the literature.

PAGE 13

13 Our second focus was an asphalt-paved inst itutional source area of approximately 450 squared meters catchment in Gainesville, Florida. Event-based hydrologic nutrient and PM data were collected. From these data, transport and partitioning were determined and concentrations of nutrients and PM were compared to similar source areas. On an event basis the median dissolved fraction of nitrogen (N) was 0.21 and 0.36 for phosphorus (P). When classified as suspended, settleable and sediment fractions, transported PM was dominated by suspended and sediment. Transported PM had a hetero-dispers e particle size distri bution (PSD) ranging from colloidal-size to coarse sand approaching 1000 micrometers. PM and PM-bound transport was mass-limited with a mean k1 value of 5.67, 5.97, and 4.58 for P, N and SSC respectively. Median values of the pa rtitioning coefficient for N and P were 41,870 and 21,753 L/kg respectively Compared to similar source areas the N, P and PM loads were higher, while the dissolved fraction and partitioning coefficient were similar. With proper treatment, rainfallrunoff could be used in a reuse irrigation application.

PAGE 14

14 CHAPTER 1 INTRODUCTION Concentrations of Pollutants According to Land Use Overall con centrations of pollutants in rainfa ll-runoff vary according to land use across the nation according to the National Stormwater Qual ity Database (NSQD) (Pitt et al. 2004). When examining which land uses have the highest and lo west concentrations of total Cd, it was found that industrial land uses exhibi t the highest a median concentr ation of 1.9 g/L among the land uses, while residential land uses display the lowe st median concentration of 0.5 g/L (Pitt et al. 2004). Additionally, the highest median total Pb and Zn concentrations came from industrial land uses at 26 and 200 g/L respectively (Pitt et al. 2004). The highest a nd lowest median total Cu concentrations came from freeway land uses at 34.7 g/L and from the open space land use at 29 g/L respectively (Pitt et al. 2004). Similarl y the open space land use ha d the lowest total Pb and Zn median concentrations of 19 and 32 g/L respectively (Pitt et al. 2004). The implication of these data that was collected from land uses including residential, commercial, industrial, institutional, highways, and open spaces is that particularly for metals, industrial land uses contribute significant amounts of pollutants to st ormwater. Besides the NSQD, a highway land use in Cincinnati, Ohio, found the EMC ranges fo r Zn, Cd, Cu, and Pb were 459 to 15,244 g/L, 5 to 11 g/L, 43 to 325 g/L, and 31 to 97 g/L for five storm events respectively (Sansalone and Buchberger, 1997). Another st udy done in New Orleans, Loui siana found that the range of EMCs for Cu, Zn, Cd, and Pb were 12.1 to 206.3 g/L, 44.1 to 985.6 g/L, 0.4 to 31.6 g/L, and 1.5 to 58.0 g/L respectively from a hi ghway catchment (Dean et al. 2005). Other pollutants were investigated by the NSQD for contributions according to land use. When examining which land uses contribute the lo west concentrations of nutrients, industrial land uses contrastly compared to heavy metals, have the lowest median concentrations for both

PAGE 15

15 total nitrogen (TN) at 2.03 mg/L and for disso lved phosphorous at 0.11 mg/L (Pitt et al. 2004). On the other hand, industrial land uses contribute the highest median concentration for nitrogen nitrite-nitrate at 0.71 mg/L, while highways cont ribute the lowest median concentration of 0.28 mg/L (Pitt et al. 2004). Fina lly, the residential land use c ontributed the highest median concentration of TN and disso lved phosphorous at 2.3 and 0.17 mg /L respectively (Pitt et al. 2004). The mean concentration from paved as phalt in Waterford, Connecticut was 47.8, 8.0, and 0.244 mg/L for TSS, TKN, and TP respectivel y (Gilbert and Clausen, 2006). Another study conducted in Phoenix, Arizona investigated the dissolved nitrogen and orthophosphate concentrations from industrial, commercial, and residential land uses. From the industrial land use the mean dissolved nitrogen and orthophos phate concentrations were 26.6 and 0.18 mg/L respectively (Hope et al. 2004). The commerc ial land use from this study found mean concentrations of 14.2 and 0.18 mg/L of dissolv ed nitrogen and orthophosphate respectively (Hope et al. 2004). Finally, the residential land use exhibited m ean concentrations of 3.4 and 0.13 mg/L of dissolved nitrog en and orthophosphate respectiv ely (Hope et al. 2004). The NSQD also investigated which land uses had the highest and low median concentrations for Total Dissolved Solids (TDS) and Total Suspended Solids (TSS). The highest and lowest median concentrations for TDS were for the industrial land use, being 92 mg/L and for the residential land use, being 72 mg/L (Pitt et al. 2004). The highest and lowest TSS median concentrations were for the highway land use at 99 mg/L, and for the residential land use at 49 mg/L (Pitt et al. 2004). Biogenic and Anthropogenic Sources It is im portant to understand the sources of thes e pollutants that become a part of rainfallrunoff during a given storm event. Both bioge nic sources and anthropoge nic sources contribute pollutants to stormwater. Biogenic sources such as vegetation contribute a substantial amount of

PAGE 16

16 Total Phosphorous (TP) and Total Nitrogen (TN) Biogenic sources can be transported into runoff because of existing design of urban land uses. For instance, if islands in an urban land use, surrounded by impervious park ing lot surfaces are at higher elevations when examined from a profile point of view, much of the nutrient debris can be transported during a storm event. Anthropogenic sources of pollution from transpor tation land use are typically from pavement wear, tire wear, engine/brake wear, settleable exha ust, and land use activities such as fertilizer or pesticide practices (Sansalone, 2001). More specif ically, brakes from a vehicle can contribute Cu and Pb, an engine of a vehicle can contribut e Zn and Cu, and tires can contribute Zn and Cd (Sansalone, 2001). Tires are estimated to cont ribute 3.0 mg/vehicle-km of Zn, while Cd can contribute 0.02 mg/vehicle-km (Sansalone, 2001). Partitioning of Pollutants Investigating the partitioning of pollutants transporte d in ra infall-runoff is of value because of greater understanding of partitioning can lead to more effectively designed treatment for rainfall-runoff. The dissolved fraction (fd) and partitioning coefficient (kd) have both been historically used to describe ex hibited partitioning by rainfall-runoff. Partitioning of metals such as Pb, Cu, Cd, and Zn has more commonly been studied from urban watersheds than nutrient partitioning. For instance, results from a highway land use in Cincinnati, Ohio indicated that Zn, Cu, and Cd predominate in the dissolved form while Pb remains predominantly particulatebound (Sansalone and Buchberger, 1997). Other ur ban highway land uses, such as one in Baton Rouge, Louisiana, found that Zn was particul ate bound, Pb was highly pa rticulate bound, and Cu and Cd partitioned more equally between the phases (Dean et al. 2005). First Flush Phenomenon There has been debate historically on how po llutants are delivered in an urban watershed. The first flush phenom enon states that a majority of pollutants are delivered in the initial portion

PAGE 17

17 of the hydrograph. Defining whether a storm event exhibits a first flush phenomenon is difficult because various researchers have used differing definitions. One study defined the first flush phenomenon as the delivery of a disproportionately large load of constituents during the early part of the runoff hydrograph (Sansalone and Cris tina, 2004). This was defined as a first flush when the dimensionless mass curve preceded the dimensionless volume curve. A history of first-flush methods was presented with a review of published first-flush methods. Additionally, other historical quantitative defini tions that are more strict for fi rst flush have included the mass ratio being at least 50 %, while the volume ratio is not more than 25 % (Wanielista and Yousef, 1993), and the mass ratio being at least 80 %, wh ile the volume ratio is not more than 20 % (Stahre and Urbonas, 1990). Numerous studies have investigat ed the pollutant first flush, es pecially metals in runoff. Studies performed in Cincinnati, Ohio from a highway land use indicated that dissolved fractions of Zn, Cd, and Cu experienced a strong first flush, while Pb exhibited a weak first flush (Sansalone and Buchberger, 1997). Another study found that tota l Cu, total Pb, and total Zn from residential, commercial, industrial, agricultural, and open-space land uses demonstrated moderate first flush patterns (Tiefenthaler et al. 2008). Other studies have investigated the tendency of nutrients such as TN and TP to experience a first flush. A study done in Chongju, South Korea with land uses that included residential, commercial, and industrial land uses found that the first flush phenomenon was greater for smaller watershed areas (Lee et al. 2002). Additi onally, the first flush for suspended solids (SS) was the greatest, as well as Total Kjeldrahl Nitr ogen (TKN) having a stronger first flush than TP (Lee et al. 2002). Another st udy conducted in Wuhan City, China found that for eight respective

PAGE 18

18 storm events that both TN and TP experienced a slight first fl ush, while the TSS demonstrated a strong first flush (Li et al. 2007). Health Effects of Pollutants Historically the Environm enta l Protection Agency (EPA) has recognized the importance of regulating both drinking water and surface water bodies in the United States (US). With the establishment of the Clean Wa ter Act (CWA) in 1972, the EPA formulated maximum allowable water quality standards for the discharge of pollutants into wate r bodies. Chronic exposure of Copper (Cu), Lead (Pb), Cadmium (C d), and Zinc (Zn) in freshwater bodies is not to be in excess of 9, 2.5, 2.2, and 120 g/L respectively (USEPA 1999). Additionally, acute exposure should not be in excess of 4.3, 13, 65, and 120 g/L for Cd, Cu, Pb, and Zn respec tively in freshwater bodies (USEPA, 1999). These values are all calc ulated using a default hardness of 100 mg/L. Similarly, the Safe Drinking Water Act (SDWA) in itiated in 1974, was established in order to protect human health with the consumption of ta p water supplied from local municipalities. The rationale of formulating drinking water standards is based on know n health effects of chronic or acute consumption of pollutants. For instance, heavy metals, such as Cu, Pb, Cd, Zn can present health impacts to humans depending on the level of exposure. In particular, Pb at high concentrations is known to hinder physical and mental development in infants and children (Needleman and Bellinger, 1991). Additionally, in adults over consumption of Pb can lead to kidney problems or high blood pr essure (Tong et al. 2000). Other metals such as Cu have known consequenc es if there is chronic or acute levels of exposure, despite the role of Cu as a trace mineral for humans. For instance, short term exposure of Cu can cause gastrointestinal disturbance, as well as nausea and vomiting (Pizzarro et al. 1999). If there is long-term exposure, liver or kidney damage can be a negative health effect (Pizzarro et al. 1999). Besides Cu causing na usea and vomiting, Cd is known to have these

PAGE 19

19 visible health effects, as well as diarrhea, muscle cramps, sali vation, sensory disturbances, liver injury, convulsions, shock, and renal failure (Ha llenbeck, 1984). Finally, Zn, while also a trace metal for humans, has known gastrointestinal e ffects when consumed in excess, as well as diarrhea, nausea, and vom iting (Walsh et al. 1994). Bioavailability of Metals The Maximum Contaminant Levels (MCLs) for Cd, Cu, and Pb are 0.005, 1.3, and 0.015 mg/L respectively, in order to ensure public safet y. Besides public safety, health of aquatic life is of concern, particularly because of pollutants th at are constituents in stormwater. Numerous studies have investigated the bioa vailability of heavy metals from stormwater to aquatic life such as fish or macroinvertebrates. Anderson et al. (2004) found that when comparing bioavailable forms of Cu, Pb, and Cd, that Pb is a possible concern in stormwater f acilities, while Cu has limited availability, and Cd is not in bioavailable forms at all. Other studies such as Mersch and Pihan (1993), observed increases in Pb, Cd, and Cu in caged zebra mussels after only 27 days of a 90-day study. Couillard et al. (1995) found th at after only 90 days of a 400 day study, that there was a 3.3-fold increase in metal a nd metal-lithionein concentrations in Anodonta grandis. Finally, Tessier et al. (1984) found that Cu and Pb levels in the freshwater mussel Elliptio complanata correlated with carbonate, Fe/Mn oxide, and or ganic/sulfide fractions from sequential extraction, and partiall y with the residual fraction a nd the total of all sediment fractions. Not only are there measurable effects on aquatic life from the bioavailability of rainfallrunoff from any urban watershed, but there are additionally distinguishable increased concentrations of heavy metals from particular land uses. Fo r instance, a study performed in Maryland that analyzed odonate inve rtebrates found that Zn concentr ations, with a mean value of 113.82 g/g, were the highest in the odonates from watersheds draining commercial land uses

PAGE 20

20 (Karouna-Renier and Sparling, 2001). Additionally, the Cu concentrations present in the odonates were higher from commercial land uses having a mean of 27.12 g/g, than highway land uses, at a mean of 20.23 g/g, or open sp ace, which exhibited a mean of 17.79 g/g land uses (Karouna-Renie r and Sparling, 2001). Consequences of Nutrient Loading The pollutant loading of heavy m etals from stormwater have negative impact on aquatic life present in waterbodies, and nut rient loading can also have an adverse influence on species quality of life. Discharge of nutrients into su rface waters can stimulate algal and plant growth, which eventually results in less oxygen bei ng available for aquatic life (Bitton, 2005). Additionally, eutrophi cation results in decreased water transp arency, increased incidence of fish kills, and loss of desirable fish species (Carpenter et al. 1998). In most cases N is more of a limiting factor than P in excessive eu trophication (Carpent er et al. 1998). Implications of Stormwater Characteristics Understanding the characteristics of resulting stormwater generated fr om urban watersheds is fundamental to implementing the proper treatment strategies. Knowledge of EMC concentrations of pollutants presen t in rainfall-runoff, such as nutrients and metals, is important, in order to realize what magnit ude of pollutants need removal. Additionally, examining the partitioning of pollutants constituen ts is of value, to understand and select unit operations or what Best Management Practices (BMP) are eff ective. Finally, investigating the first flush phenomenon of pollutants is necessary to dete rmine whether volumetric treatment strategies have been properly implemented. Once agai n, the overall goal of studying pollutants in stormwater is to provide the foundation for effectiv e removal to preserve aquatic life, as well as protect the general public from contaminated waterbodies.

PAGE 21

21 CHAPTER 2 PARTITIONING AND TRANSPORT OF METALS IN RAINFALLRUNOFF FROM AN INDUSTRIAL SOURCE AREA WATERSHED Introduction The transpo rt and phase-partitioning of constitu ents in runoff on the surface of the built environment are coupled phenomena that are influe nced by available const ituent load, solubility, hydrologic phenomena and anthropogenic alterations to the watershed surfac e. Publications over the last several decades have demonstrated that such anthropogenic alterations, such as the degree of imperviousness result in rainfa ll-runoff modification (H uber 1993); and impact transport of pollutant loads (Morisawa et al. 1979, Heaney and Huber 1984; Hamilton and Harrison 1991; Arnold and Gibbons 1996). Specif ically, with respect to rainfall-runoff modification, these publications have identified increased runoff volume, and peak flow. Given the coupling between hydrology and water chemistr y, modifications to tr ansport processes and loads are a consequence. While at what de minimis threshold impervious levels have a demonstrable impact is variable studies have illustrated the clear imp act of imperviousness on hydrology and load transport (Lee and Heaney 2003, Rose and Peters 2001). The impervious %age of a particular site can vary depending on its type of land use. Previous investigation indicates that industrial sites typically have a mean imperviousness ranging from 40 to 60 %, while other types of land uses such as commercial have a higher mean imperviousness that can range from 80 to 90 %, and single-family re sidential land uses contain a lower mean imperviousness ranging from 15 to 40 % (Brabec et al., 2002). Not only is the aerial extent of imperviousness important, but the types of imperv ious surfaces influence the pollutants that are transported in runoff. For example, in a Ma ryland study on urban commercial land use runoff, the main source of Pb and Zn were building sidi ng, while Cu and Cd were attributed to roof surfaces (Davis et al. 2001). Other studies indica te that leaching rates of Cu and Zn from metal

PAGE 22

22 roof cladding are 1.1g/m2 per year and 3.0g/m2 per year respectively (Korenromp and Hollander, 1999). The study site is a formerly-act ive industrial site. Industrial source area watersheds that include manufacturing or handling of chemical products or highway watersheds can contain legacy loads of chemicals, for ex ample in surficial soils that can be leached or transported during rainfall-runoff events or over extended periods of time. For example, while Pb has been phased out as an octane-enhancer in motor vehicle fuel s since the 1980s and banned since 1996, the use of Pb since the 1930s has resulted in legacy contamination in soils of highway, urban and industrial land uses (Sansalone and Glenn, 2005; Turer et al., 2001; Ayres, 1992). Even though tetraethyl lead ( CH3CH2)4Pb has been largely eliminated in the USA, Pb is estimated to have a lifetime of 3.0 x 103 years on land and 99.0 x 106 years in sediments respectively (Ayres, 1992). Therefore, despite the wide spread elimination of ( CH3CH2)4Pb manufacturing, Pb as a legacy pollutant in surficial soils a nd urban surfaces, can be leached or scoured in rainfall-runoff processes. Ag, Cu, and Zn utilized as industr ial catalysts, reagents, and in motor vehicle components such as brake linings (Cu) and metal galvanizing processes (Zn) are leached to and from soils by rainfall-runoff processes or transpor ted in runoff (Ayres, 1992; Turer et al. 2001). With respect to partitioning to or from soils or PM entrained in runoff, there are three primary mechanisms of bonding preferences; cova lence, electrostatics, and hydrolysis followed by a sorption process; utilized to explain partitio ning. Of these, hydrolysis has been used to explain the bonding preference to PM or surficial soils for aqueous metals in runoff where Pb > Cu > Zn > Co > Ni > Cd > Mn > Ag (McBri de, 1994; Baes, and Messmer, 1976; Burgess, 1978; Sansalone and Glenn 2007) Partit ioning results for small highly impervious urban catchments illustrate that Pb is preferentially bound to entrained PM followed by Cu and Zn. In addition to

PAGE 23

23 metal chemistry, aqueous chemistry, PM, and reside nce time influence partitioning (Dean et al., 2005). Objectives Our study had num erous objectives for partitioni ng and transport of metals, as well as transport of PM in rainfall-runoff from an industrial source area watershed. The first objective was to classify the hydrologic tran sport of dissolved and particulate metal and PM phases for the industrial source area watershed. The second ob jective was to quantify the phase partitioning of metals for this watershed, through analysis of the dissolved fraction and the partitioning coefficient. The third objective was to compare metal concentrations from this watershed to similar land use watershed or source area waters heds with metal loadings. Finally, it is important to distinguish whether th e rainfall-runoff generated from th is site has the capability of being treated appropriately to be used as a reso urce for other industries th at may need water for cooling towers or have irrigation demands. Background The imm ediate receiving water for discharges from the study watershed of 73 ha is the Dominguez Channel. The Domingu ez Channel watershed is 293 km2 and dominated by urban land uses that include dense residential (41%), industrial (17%), commercial and public (14%), transportation (13%) with a combination of all other land uses at 15%. The Dominquez Channel watershed has an imperviousness of 59% (W eston, 2005). The mean precipitation for the watershed ranged from 7 to 70 mm based on years of record from 1959 to 2007. (U.S.D.C., 2008). This translated to a range of annua l runoff from the watershed of 1,800,000 m3 to 3,500,000 m3. Studies have indicated that for the years of 1994 through 2005, the COC (constituents of concern), that exceed WQO (Water Quality Objectives), include total Pb, Cu, and Zn

PAGE 24

24 concentration in the Dominguez Channel (W eston, 2005). For inst ance, the total Pb concentration increased in the Dominguez Cha nnel from 2.5 mg/L in 2001-2002 to 11.0 mg/L in 2004-2005 (Weston, 2005). The mean exceedence ratios (COC/WQO) for total Cu, Pb, and Zn were 4.1, 2.2, and 1.2 respectively (Weston, 2005). Not only were the mean exceedance ratios for total metals of concern, but the ratios for the dissolved phase of Cu and Pb were 1.7 and 1.2 respectively (Weston, 2005). The study also indi cated that of the dissolved and particulate phases only dissolved Zn illustrated a mass-limited behavior. At a 95% confidence level the mean annual load estimates of total Cu, Pb and Zn to the Dominguez Channel are 0.4 2.4, 0.2 1.1, 2.1 11.0 metric tons per year, respectively (T ienfenthaler et al., 2008). For the study watershed the permit limits, orig inally based on the California Toxics Rule (CTR), are 5.8, 221, 95, and 2.2 g/L for total Cu, Pb, Zn, and Ag; 30 mg/L for TSS, 75 NTU for turbidity, and 6.09.0 for pH, respectively. (US EPA Federal Register 65(97), 2000). These results can be compared to other data bases. Results from the National Stormwater Quality Database (NSQD) indicated that fo r industrial land uses, runoff event mean concentrations for total Cd, Cu, Pb, and Zn were 435 g/L, 455 /L, 452 g/L, and 473 g/L respectively. Additionally, the dissolved EMC for Cd, Cu, Pb, and Zn were 42 g/L, 42g/L, 51 g/L, and 42 g/L respectively (NSQD, 2003). Anot her study of an industrial source watershed in Australia found that the range EMCs and standard deviations for Zn, Cu, and Pb were 0.1500 g/L, 0.140 g/L, and 0.13.0 g/L respectively (Herngre n, 2005). A study of a paved highway linking major industrial areas in Bat on Rouge reported EMCs of 12.1 to 206.3 g/L for Cu, 1.5 to 58.0 g/L for Pb; and 44.1 to 985.6 g/L for Zn (Dean et al., 2005). Another study of four industrial sites in Los Angeles reported EMCs for Cu, Pb, Zn, and TSS were 70 g/L, 24.1 g/L, 599.1 g/L, and 92.0 mg/L, respectively (Tiefe nthaler et al. 2008). Results from an

PAGE 25

25 industrial zone in Ashdod, Israel reported EM Cs and standard devi ations of 65 (34) g/L, 8 (2) g/L, and 25 (16) g/ L for Zn, Pb, and Cu respectively (Asaf, 2004). In addition to pollutant concentrations and loads, there is widespread interest in quantifying the transport of pollutants on a volumetric flow basis, irresp ective of the pollutant phase or index. A commonly assumed concept of pollutant transport, the first flush, was initially described as a first foul flush almost a century ago to classify the transport of organic equine fecal matter when horses were the primary mode of urban transporta tion (Metcalf and Eddy, 1916). This concept is the genesis of more r ecent first flush reincarnations such as a water quality treatment volume (WQV). Such concepts have an underlying assumption that hydrology and chemistry (or pollutant load) are coupled with a mass limited behavior, a behavior that represents one limit of transport behavior (Sheng et al., 2008). Such concepts are intended to identify flow volume-based criteria for unit operati ons intended to capture and treat runoff. While a variety of analysis methods have been used to quantify pollutant transport and treatment volume requirements; ranging from hist orical first flush methods to current WQV, these methods can be summarized as mass-based analysis, concentration-based analysis and empirical methods (Sansalone and Cristina, 2004). The plethora of literature on the topic points to the continued interest in a simple volumetric basis to evaluate pollutant transport and treatment requirement for runoff (Stahre a nd Urbonas, 1990; Wanielis ta and Yousef, 1993; Vorreiter and Hickey, 1994; Gupta and Saul, 19 96; Cristina and Sans alone 2003). This study utilizes a physically-bas ed classification of tran sport as either a firstorder exponential transport of mass (mass-limited) or a zero-order transport of mass (flow-limited). A statistical and physical derivation of these transport limits have been previously developed (Sheng et al., 2008).

PAGE 26

26 As the watershed becomes larger and more hydrau lically-complex, previous studies indicate that a mass limited behavior is attenuate d (Sansalone and Cristina 2004). Site Description The industrial land use site is com prised of approximately 73 ha which is divided into two catchments based in the existing surface and stor m pipe sewer drainage systems within the watershed. These two catchments are identified as the East catchment, which has 173 ha, and the West catchment, which has 212 ha. A plan vi ew of the watershed and drainage network is illustrated in Figure 2-1. The three major surface water drainage ditches are identified as the East Major Interior Collector (EMIC), the West Major Interior Coll ector (WMIC) which combine at the North Ditch combine before eventually discharging into the Dominguez Channel. The EMIC and WMIC control points, identified in Fi gure 2-1, are the locations of fl ow measurements and sampling. The imperviousness of the East and West catchme nts were 40 and 43 %, for respectively. The composition of the impervious surfaces includes pa vement, concrete structures, tanks and roof structures. The pervious surfaces comprised of anthropogenically-altered soils with vegetative scrub including tumbleweed include Ramona Lo am, Chino Silt Loam, and Hanford Fine Sandy Loam. On an aerial basis Hanford Fine Sandy Loam occupies approximately 73 %, while the Chino Silt Loam and the Ramona Silt Loam occupies 19 and 8 % of the watershed respectively. The Hanford Fine Sandy Loam is slightly alkali ne, and has a clay content averaging 6 to 18 % (NRCS, 1997). The surficial soils are reasonab ly well drained, and have a moderate hydraulic conductivity (NRCS, 1997). The ditches that coll ect the rainfall-runoff from these catchments are shallow, ranging from 0.25-0.33 meters. The nominal surface drainage slopes for the East and West catchments are 0.1% and 0.05% respectively.

PAGE 27

27 Methodology Rainfall-Runoff Measurements and Sample Collection During the 2002-2003 wet weather season, four rain fall-runoff events were collected out of seven rainfall-runoff events for that season. Rainfall was m easured and logged with a tipping bucket rain gauge and data logger in 0.254 mm in crements. The rain gauge was located in the study watershed. Hydraulic and sa mpling locations were constructe d at discharge control points at the downstream end of the EMIC and WMIC. At each v-notch weir control point, depth measurements were measured at 1 minute interval s with ultrasonic sensors, logged to the data logger and manually checked with a staff gage at each sampling time. Samples were collected manually from the free overflow at each 120 degree sharp-crested v-notch weirs. V-notch weir designs followed the specificati ons of Bos (1989). Once accurate hyetographs and hydrographs were created for the four respective storm even ts, the lag time between the hyetograph and the hydrograph, i, was calculated. Turbidity (NTU), pH (s.u.), redox poten tial (mV), and specific conductivity ( S/cm) were measured on site with meter/electrode combinations calibrated prior to each event. To ensure quality assurance and quality control of the collected samples, sampling equipment was decontaminated after each storm event with ten % trace-metal grade nitric acid and de-ionized water. All other equipment was decontaminat ed with the use of both non-phosphate detergent and de-ionized water wash. Samples were ge nerally collected at 20 minute intervals until the peak stormwater discharge was sampled, and a minimum of twenty samples was required for each storm event. If twenty-four samples were collected, and the peak discharge flow had already been captured, the sampling inte rval was extended to 60 minutes.

PAGE 28

28 Analytical Procedures Imm ediately following each event, samples were transported to the laboratory within one hour for water chemistry analysis. Alkalin ity was measured by titration (Method 2320 B) (APHA 1998). Conductivity and TDS we re measured using a Hydac sensor variable function conductivity meter. A th ree-point standard curve for TDS wa s developed for calibration prior to each event analysis. TDS was used to approximate a mass balance on the dissolved constituents of storm water. Redox potential, pH, and temperature were measured in duplicate using a Hydac sensor combination electrode. Redox potential (hydrogen-based) was determined using a potentiometer that records the potential differe nce between a platinum electrode and reference electrode immersed in a soluti on (Kamon et al. 2001). Redox poten tial was verified using an oxidation-reduction potentia l (ORP) standard solution of +435 mV at 25 C (Method 2580A) (APHA 1998). The pH meter, on the Hydac sensor, was calibrated using a three-point calibration curve with reference units at a pH of 4, 7, and 10. Total suspended solids and volatile suspended solids concentrations were analyzed for each sample following Method 160.2 and 160.4 respectively (APHA 1998). Samples were vigorously mixed and the entire suspended solids sample volume immediately passe d through a pre-weighed nominal 1 m glass fiber filter (APHA 1998). Within 12 hours from the time of sampling, samples were fractionated into their dissolved and particulate phases to minimize time dependant changes in metal partitioning (Sansalone et al. 1997). Fract ionation was accomplished by passing 60 mL of sample through a 0.45 m membrane filter. For an operational defi nition, dissolved constituents are defined as those with nominal diameters less than 0.45 m (APHA 1998). Upon fractionation, the filtrate was acidified to a pH of 2 using trace metal HNO3. Particulate matter collected on the

PAGE 29

29 membrane filter was acid digested following Standard Methods 3030D (APHA 1998). Dissolved and particulate metal analysis was co nducted with an inductively coupled plasma spectrometer (ICP-MS). A five point standard cu rve was run prior to sample analysis. Quality control, blanks, and mass balance checks were maintained throughout the cycle. A standard solution was checked for every 20 samples to insure standard checks were within + 5%. Sample analysis of dissolved and particul ate fractions was utilized to determine the partitioning coefficient, Kd (L/kg), and the dissolved fraction, or fd. EMCs were calculated. (2-1) In this equation, EMC is the flow weighted average concentrati on for the entire event, M is the total mass of constituent phase over entire ev ent duration, V is the to tal volume of flow over entire event duration, c(t) is the time variable di ssolved or particulate boun d concentration, q(t) is the time variable flow, and t is time (Hube r, 1993). Once the partitioning and EMCs were determined, cumulative mass of a constituent phase as a function of volume, or M(V) curve, was developed. After the M(V) curve is formulated, mass-limited vs. flow-limited behavior can be determined for each of the storm events through th e utilization of the firs t-order, mass-limited and zero-order flow-limited equations (Sheng et al. 2008). (2-2) (2-3) Mt is the cumulative mass delivered, Vt is the cumulative volume, M0 is the constituent mass on the surface at the beginning of the rainfall-runoff event, k1 is the first order coefficient and k0 is the zero-order coefficient. Results Hydrology Using the methodology described above, these fo ur respective storm events could be successfully characterized based on their exhibi ted hydrologic behavior. The west catchment

PAGE 30

30 was significantly larger, being 348,042 m2 as opposed to 279,243 m2 in the east catchment. The width of the watershed is approximately 675 m in the west catchment, in contrast to 350 m for the east catchment. These differences demonstr ably influence the hydro logic transport of the rainfall-runoff, as well as the hydrologic indices for these catchments (Table 2-1). For example, tmean represents the average residence time of the rainfall-runoff in the east catchment and west catchment watersheds before it reaches the outfall destination of each of the respective ditches in each catchment. The west catchm ent, since it is a larger watershed, and has a relative longer flow path, has a greater tmean of 2.53 hours, over the east catchment tmean of 2.41 hours. A number of observations regarding these four storm events illustrate rainfall-runoff transport. For example, the 20 December 2002 storm was the least intense storm of the f our events monitored, while the 12 February 2003 event was the most inte nse event. The differences in the intensity and total volume (Figure 2-2) that an events experience has measurable effects on whether the delivery of pollutants will be either mass-limited of flow-limited. The volumetric runoff coefficient is C (Table 21) but the runoff coefficient (Figure 2-3) is a function of time for each storm event. The event-based runoff coefficients range from approximately 0.2-0.5 in the west catchment after reaching the asymptotic level, while the overall runoff coefficients range from approximately 0.3-0.6 in the east catchment. M(V) Curves: Metals The plots of the cumulative metal mass as a function of cumulative runoff volume (Figure 2-4) reveal mass-limited or flow-l imited behavior of the particul ate-bound and dissolved fraction of Cu, Pb, Ag, and Zn can be accomplished for both the east catchment and the west catchment. Additionally, there is a total mass for the particulate-bound and dissolved fraction of Cu, Pb, Ag, and Zn for each respective storm event (Table 2-2). Additionally, there are fitted k0 and k1 values for the flow-limited and mass-limited storm events (Table 2-3). Those storms where

PAGE 31

31 pollutants such as Ag were not detectable are indicated as ND. The decision to focus on Cu, Pb, Ag, and Zn was determined because these me tals all had EMCs that exceeded their permit limits. When analyzing overall trends for Cu, Pb, Ag, and Zn, in both the west catchment and the east catchment, metals tend to have mass-limited behavior for particulate-bound metals and flow-limited behavior for the dissolved fraction of metals. The first order and zero order curves are relatively similar for Cu, Pb, and Zn, but significantly different than Ag. While the masslimited curves for Cu, Pb, and Zn are relatively consiste nt, yielding high R2 values when fitted, the Ag mass-limited curves fluctuate immensel y forming a mass-limited curve that sometimes reaches an asymptote for a significant volume period, before establishing a mass-limited trend again until it reaches the Vmax. Explanation for the differing be havior of Ag, may be that Ag tends to follow the hydrograph of intense runoff periods more closely than do other metals. M(V) Curves: TDS and TSS Not only is it im portant to monitor the mass-lim ited vs. flow-limited behavior of metals, in order to better understand how to effectively treat these pollutants, but it is important to examine other pollutants for their behavi or as well. There are trends for the TDS and TSS for both catchments for each respective storm event (Figure 2-5). There is a total mass for the TDS and TSS for each storm event (Table 2-4) Additionally, there are fitted k0 and k1 values for the flow-limited and mass-limited storm events (Tab le 2-5). When analyzing the TDS, it can be determined that TDS follows a flow-limited behavior in both the east catchment and west catchment, for all storm events except for the 12 February 2003 event that follows a strong masslimited behavior. This result is consistent with the dissolved fraction of metals exhibited masslimited behaviors in the east catchment dur ing the 12 February 2003 storm event.

PAGE 32

32 Unlike TDS, the TSS, tends to exhibit more mass-limited behavior for all of the storm events, except in the west catchment for th e 20 December 2002 storm event, which is flowlimited. The TSS of the 20 December 2002 event in the west catchment corresponds similarly to the particulate-bound metals in the west catchment that were flow-limited. The remainder of the results of the TSS being mass-limited is expecte d, because if the metals that are particulatebound are demonstrating mass-limited behaviors, one would expect the TSS behavior to similarly be mass-limited. When comparing thes e results to previous research, they are consistent with results such as TSS being ma ss-limited in 66 % out of a total of 80 events (Bertrand-Krajewski, 1998). The TDS and TSS results for this study are different than the results produced in Cincinnati that stated that the TDS experienced a more rapid first flush than TSS (Sansalone and Buchberger, 1997). M(V) Curves: Other Pollutants Besides exam ining the mass-limited and flow-limited behavior of metals and TDS and TSS, the final pollutants analyzed for th is study were the DOC TOC, ALK as CaCO3, and hardness as CaCO3 (Figure 2-6) There is a total mass for th e DOC, TOC, ALK, and hardness for each respective storm event (Table 2-4). In both the east catchment and west catchment, flow-limited behavior was mostly exhibited for all storm events except for the 12 February 2003 event in the west catchment, which displays st rong mass-limited behavior for DOC, TOC, and hardness as CaCO3. This storm is consistent with pr evious results in this study for other pollutants like TDS, Cu, Pb, and Zn. The results in this study for TOC are similar to the results produced by Geiger in 1987, which were flow-limited for the median curves, but not for TSS which was flow-limited for the median curves (Geiger, 1987).

PAGE 33

33 EMC Utilizing th e methodology described above, the EMC was calculated for relevant rainfallrunoff pollutants. There are particulate-bound an d dissolved fraction EMCs for Cu, Pb, Ag, and Zn (Table 2-6). Additionally, there are EMCs and EMVs for the remaining pollutants such as TDS, TSS, TOC, DOC, Alkalinity, hardness, con ductivity, turbidity, pH and Redox (Table 2-7). Low alkalinity values tend to correspond with a higher affinity for the dissolved fraction of pollutants in rainfall-runoff. Similarly, when the pH of the rainfall-runoff is only slightly acidic, or nearly neutral, there is a higher affinity for the dissolved fraction. In this case, for all four respective storm events have low EMC Alkalinit y concentrations since they are less than 31 mg/L, and the EMV for the pH is nearly neutral/slightly acidic by ranging from 6.6-7.8. Because values for parameters such as conduc tivity, pH, turbidity, and redox are influential to partitioning and transport thr oughout the entire storm, it is not sufficient to solely summarize these parameters with EMCs and EMVs. To further understand how these parameters can influence the overall par titioning behavior, the value of thes e parameters throughout the entire 20 December 2002 and 12 February 2003 storm events (F igure 2-7), while similarly there are results for the 24 February 2003 and 15 March 2003 storm even ts (Figure 2-8). The pH stays relatively constant throughout all of the storm ev ents samples ranging between 6.0-8.0. It is important to compare the metal EMC resu lts for the industrial source at hand to other urban sites that have previous ly been studied, to determine whether the magnitude of the pollutant concentrations are sim ilar or significantly different. These catchments pollutants of Cu, Pb, Zn, and Cd compare to rainfall-runoff that was analyzed from other sites (Figure 2-9). Three industrial studies are included in comparison to this site, includi ng studies performed in Milwaukee, WI; Lund, Sweden; and another study pe rformed in Greater Los Angeles, CA. The

PAGE 34

34 remaining two studies are from catchments off of major highways, in Baton Rouge, LA and Cincinnati, OH, which to an exte nt are similar to the impervious surfaces that comprise this watershed. Fd and Kd Not only are the EMCs of pollutants such as metals, important in determining the partitioning of rainfall-runoff, but fd and kd reflect the differences in partitioning from each type of metal as well. There are box plots of both fd and kd for Cu, Pb, and Zn metals (Figure 2-10). Out of the three metals, Pb has the least affinity for the dissolved fraction in both catchments, typically ranging from approximately 0.0 to 0.2 for fd. Zn has the highest range of variability of the metals for fd ranging from approximately 0.0 to 0.9, while Cu partitioned more equally between phases having a range between 0.1 to 0.8. The results for Pb are expected considering that Pb, as mentioned previously is most likely to undergo hydrolysis out of all of the metals. When examining the results for the other meta ls, it is important to consider how other influential factors in partitioning such as residence time, alkalinity, and pH affected these results. Since there is low alkalinity and neutral pH exhi bited in all four respective storm events, one would expect the other metals to still remain in the dissolved fraction. But when examining the mean residence times for the east catchment and west catchment catchments, of 2.41 and 2.53 hours respectively, it is known that these are not s hort residence times. Because these are longer residence times, there is more opportunity for th e dissolved fraction of pollutants to undergo sorption to the particulate-bound phase. Ultimately, the presence of low ALK, neutral pH, and longer residence times, produced results for Cu and Zn being more equa lly partitioned between the two phases. The Kd plots, which are more representative of the respective metals tendency to be particulate-bound, reveal similarly the highest Kd values for Pb, on the order of 1x105 to 1x106

PAGE 35

35 L/kg, where as the other metals Cu and Zn are on orders of 1x104 to 1x105 L/kg. The variation in the Kd values between each storm event and catchment for Cu and Zn ultimately result in it being difficult to state that there is a significant difference in partitioning between the two metals. Once again, it is important that these results do no t exist in isolation. Therefore, there is a box plot revealing how the fd and kd values for this industrial source watershed compare to the Baton Rouge, LA and Cincinnati, OH urban sites mentioned previously in the EMC comparison (Figure 2-11). Overall, the results for fd and kd in this study are most similar to the Baton Rouge, LA site. The Cincinnati site has a si gnificantly higher tendenc y to be particulate bound compared to the other two urban sites for Cu, Pb, Zn, and Cd. Implications/Discussion There are significant implications of the results of this study for stormwater management practices. One of the important issues of c oncern with rainfall-runo ff from industrial source watersheds, such as the east catchment and west catchment in Los Angeles County draining into the Dominguez Channel, is measuring the bioavailability of the constituents such as metals. Before measuring the bioavailability of the metals that affect organisms such as fish in the water body, the immediate impacts of pollu tion with contaminants such as metals can be measured by examining the benthic macroinvertebrate communities. One of most useful tools in measuring this effect is with the CFGs Southern California Index of Biotic Integrity. Scores can range on this index from 0 to 70 where lower scores are rate d as very poor and higher scores are rated as very good. The Dominguez Channel monitoring stat ion, that is located in the main stem, received scores of 3 in 2003 and 6 in 2004 (Weston, 2005). Therefore, these benthic

PAGE 36

36 macroinvertebrate communities are consistently negatively impacted by receiving storm water with pollutants such as metals. Because there is risk involved of not only the dissolved portion of rainfall-runoff being absorbed by marine life, but the particul ate-bound portion of ra infall-runoff becoming bioavailable after length s of time, it is important to continue to analyze the partitioning that exists for all pollutants. If the differences in partitioning between metals such as Pb, Cu, and Zn, are better understood through further research, the proper treatment st rategies can be initiated to properly treat each part of the pollution spectrum, before it reach es outfalls where marine life reside. Additionally, if proper treatment is initiate d, this rainfall-runoff can be used instead as reuse water to supply other industrie s that need water to supply cooling towers or have irrigation demands. Because previous research indicates that th e first-flush phenomenon is not exhibited in larger watersheds, but primarily in smaller watersheds, it is crucial to implement technology in BMPs that treat both the dissolv ed and particulate-bound fraction for the entire length of the storm event (Tiefenthaler et al., 2008). This treatment is not currently implemented by most BMPs, so further research is n eeded to conclusively prove that pollutants are delivered evenly throughout the entire storm event. Conclusions Our study reveals that even in cases where there are industrial sites that are no longer being used for activities such as oil refining signifi cant amounts of pollutants st ill exist years later in rainfall-runoff, and cannot be ignored. At this pa rticular industrial source watershed, the metals including Pb, Cu, Ag, and Zn, that still exceed permit limits, are originating from the leaching of metals from impervious roofs and from parki ng lot surfaces, and are ul timately contributing to the pollution of the Dominguez Channel.

PAGE 37

37 The four respective storm events captured for this study, varied sign ificantly in rainfall intensities and total runoff flow volumes. Because of the range of low intensity, low flow volume events and high intensity, high flow volum e events, in combination with the hydrological conditions, there were varying transport of the rainfa ll-runoff at the site. Factors such as length of flow path, residence time, and previous dry ho urs all had measurable effects in the resulting transport of the rainfall-runoff. Besides being able to effectiv ely measure the transport of ra infall-runoff at the industrial site, the partitioning was distinguishable as well While the Pb presen t in the rainfall-runoff present is dominant in a partic ulate-bound state, having a mean fd value of 0.036, other metals such as Cu and Zn are more equally partitio ned between the dissolved and particulate-bound fraction, indicating that both phases must be treated effectively a nd in its entirety. Treatment strategies for this site may be difficult to de velop, because for most of the dissolved metal pollutants in both catchments, as well as TDS, with a median k0 value of 0.979, TOC, DOC, and hardness, tend to exhibit flow-lim ited behavior, and consequently do not experience a first-flush. Since these large watersheds do not exhibit the first flush phenomenon in their rainfall-runoff, the early part of the spectrum of pollutants cannot be the only fo cus for treatment. One of the important outcomes of this study was to examine whether the industrial sites rainfall-runoff partitioned and exhibited transport si milar to other urban sites. When analyzing the EMCs from two highway catchment sites, as well as three other industrial si tes, it is clear that the EMC for Cu, Pb, Cd, and Zn, with mean concentrations of 55.9, 181.4, 0.64, and 143.7 can range in significant magnitudes. When investigati ng further into the partitioning of these metals at these urban sites, the fd for Pb ranges between 0.0-0.2, whil e the Cu, Cd, and Zn are largely more variant between the sites.

PAGE 38

38 Figure 2-1 Plan view of watershed located in Great er Los Angeles with delineation of WMIC and EMIC and drainage flow paths 823 m 671 m 286 m North Ditch (effluent discharge) East Catchment 1 1 EMIC Control Point 2 WMIC Control Point 3 North Control Point 2 3 West Catchment

PAGE 39

39 05101520Runoff (L/s) 0 50 100 150 200 Cumulative Runoff (L) 0 1e+4 2e+4 3e+4 4e+4 5e+4 6e+4 Rainfall (mm/hr) 0 20 40 60 80 010203040 05101520 Rainfall (mm/hr) 0 20 40 60 010203040Runoff (L/s) 0 500 1000 1500 2000 2500 3000 Cumulative Runoff ( L ) 0.0 5.0e+4 1.0e+5 1.5e+5 2.0e+5 2.5e+5 3.0e+5 3.5e+5 EMIC Elapsed Time (hours) 051015202530Runoff (L/s) 0 100 200 300 400 Cumulative Runoff (L) 0.0 2.0e+4 4.0e+4 6.0e+4 8.0e+4 1.0e+5 1.2e+5 1.4e+5 1.6e+5 Rainfall (mm/hr) 0 10 20 30 40 WMIC Elapsed Time (hours) 051015202530 02468101214Runoff (L/s) 0 20 40 60 80 100 Cumulative Runoff (L) 0.0 2.0e+3 4.0e+3 6.0e+3 8.0e+3 1.0e+4 1.2e+4 1.4e+4 1.6e+4 Rainfall (mm/hr) 0 5 10 15 20 02468101214 15 March 2003 15 March 2003 24 February 2003 24 February 2003 12 February 2003 12 February 2003 20 December 2003 20 December 2003 Figure 2-2 Hydrology hyetographs, cumulative and incr emental hydrographs as a function of time for the storm events. Time 0 represents start of rainfall.

PAGE 40

40 WMIC Elapsed Time (hours) 05101520Rainfall-Runoff Coefficient, C 0.0 0.2 0.4 0.6 EMIC Elapsed Time (hours) 05101520 20 December 2002 15 March 2003 24 February 2003 12 February 2003 15 March 2003 20 December 2002 24 February 2003 12 February 2003 Figure 2-3 Runoff coefficients for storm events for both WMIC and EMIC

PAGE 41

41 WMIC Cu Cumulative M/M max 0.0 0.2 0.4 0.6 0.8 1.0 20 Dec 2002 24 Feb 2003 15 March 2003 12 Feb 2003 WMIC Pb 20 Dec 2002 24 Feb 2003 15 March 2003 12 Feb 2003 WMIC Zn WMIC Ag Cumulative M/M max 0.0 0.2 0.4 0.6 0.8 1.0 EMIC Pb EMIC ZnCumulative V/V max 0.00.20.40.60.81.0 EMIC AgCumulative V/V max 0.00.20.40.60.81.0 Cumulative M/M max 0.0 0.2 0.4 0.6 0.8 1.0 EMIC Cu Cumulative M/M max 0.0 0.2 0.4 0.6 0.8 1.0 Particulate Dissolved Figure 2-4. Plots of cumulative mass fractions as a function of cumulative runoff volume for metals

PAGE 42

42 WMIC TDS Cumulative M/M max 0.0 0.2 0.4 0.6 0.8 1.0 20 December 2002 24 February 2003 15 March 2003 12 February 2003 WMIC TSS EMIC TDSCumulative V/V max 0.00.20.40.60.81.0 Cumulative M/M max 0.0 0.2 0.4 0.6 0.8 1.0 EMIC TSSCumulative V/V max 0.00.20.40.60.81.0 Figure 2-5. Plots of cumulative mass fractions as a function of cumulative runoff volume for TDS and TSS

PAGE 43

43 Figure 2-6 Plots of cumulative ma ss fractions as a function of cumulative runoff volume for DOC, TOC, ALK as CaCO3, and Hardness as CaCO3 WMIC DOC Cumulative M/Mmax 0.0 0.2 0.4 0.6 0.8 1.0 20 December 2002 24 February 2003 15 March 2003 12 February 2003 WMIC ALK as CaCO 3 0.0 0.2 0.4 0.6 0.8 1.0 EMIC ALK as CaCO 3 WMICTOC 0.0 0.2 0.4 0.6 0.8 1.0 EMIC DOC Cumulative M/Mmax Cumulative V/V max 0.00.20.40.60.81.0 0.0 0.2 0.4 0.6 0.8 1.0 WMICHardness as CaCO 3Cumulative M/Mmax Cumulative M/Mmax Cumulative V/V max EMIC TOC EMIC Hardness as CaCO 3 0.00.20.40.60.81.0

PAGE 44

44 0.00.20.40.60.81.0 Redox (mV) 0 500 1000 1500 2000 20 December 2002 Conductivity ( mS /cm) 0 500 1000 1500 2000 2500 Hydrograph Conductivity Turbidity Turbidity (NTU) 0 200 400 600 800 pH 0 2 4 6 8 10 Hydrograph pH Redox Redox (mV) 0 100 200 300 400 500 0 200 400 600 800 1000 1200 Turbidity (NTU) 0 200 400 600 800 1000 1200 0.00.20.40.60.81.0 pH 0 2 4 6 8 10 WMIC WMIC WMIC WMIC EMIC EMIC EMIC EMIC t/t max t/t max t max =15.89 hours t max =33.42 hours 12 February 2003 t max =33.70 hours t max =12.20 hoursConductivity ( mS /cm) Figure 2-7. Conductivity, pH, turb idity, and redox incremental pl ots as a function of time for the 20 December 2002 and 12 February 2003 storm events

PAGE 45

45 0 200 400 600 800 Conductivity (mS/cm) 0 1000 2000 3000 4000 5000 Hydrograph Conductivity Turbidity Conductivity (mS/cm) 0 200 400 600 800 1000 Turbidity (NTU) 0 200 400 600 800 1000 1200 1400 t/t max 0.00.20.40.60.81.0 pH 0 2 4 6 8 10 t/t max 0.00.20.40.60.81.0 Redox Potential (mv) 0 50 100 150 200 250 300 350 24 February 2003 pH 0 2 4 6 8 10 Hydrograph pH Redox 15 March 2003 Redox Potential (mv) 0 50 100 150 200 250 300 350 Turbidity (NTU)WMIC WMIC WMIC WMIC EMIC EMIC EMIC EMIC t max =18.58 hours t max =33.60 hours t max =15.60 hours t max =24.70 hours Figure 2-8. Conductivity, pH, turb idity, and redox incremental pl ots as a function of time for the 24 February 2003 and 15 March 2003 storm Events

PAGE 46

46 Total Cu [ g/L] 0.01 0.1 1 10 100 Total Pb [ g/L] 0.01 0.1 1 10 100 1000 Total Zn [ g/L] 0.1 1 10 100 1000 Present Study, CA Total Cd [ g/L] 0.001 0.01 0.1 1 10 Baton Rouge, LA Cincinnati, OH Milwaukee, WI Lund, Sweden* Los Angeles, CA* Present Study, CA TSS [ g/L] 0.1 1 10 100 1000 Baton Rouge, LA Cincinnati, OH Milwaukee, WI Lund, Sweden Los Angeles, CA Figure 2-9 Comparison of the sample concentrations of Cu, Zn, Pb, and Cd, as well as TSS at the greater Los Angeles site to other urban and industrial sites Data Not Available Mean 95th %ile 5th %ile 90th %ile 10th %ile Median 25th %iles 75th %iles

PAGE 47

47 Figure 2-10 Dissolved fraction and partitioning co efficient statistics for metals Cu f d 0.0 0.2 0.4 0.6 0.8 1.0 WMIC EMIC 20 Dec 02 12 Feb 03 24 Feb 03 15 March 03 Cu k d (L/kg) 1.0e+2 1.0e+3 1.0e+4 1.0e+5 Pb f d 0.0 0.2 0.4 0.6 0.8 1.0 Zn Storm Event f d 0.0 0.2 0.4 0.6 0.8 1.0 Pb k d (L/kg) 1e+4 1e+5 1e+6 Storm Event k d (L/kg) 1e+1 1e+2 1e+3 1e+4 1e+5 Zn20 Dec 02 12 Feb 03 24 Feb 03 15 March 03

PAGE 48

48 f d : Cu f d 0.0 0.2 0.4 0.6 0.8 1.0 f d : Zn f d 0.0 0.2 0.4 0.6 0.8 1.0 f d : Pb f d 0.0 0.2 0.4 0.6 0.8 1.0 f d : Cd Baton Rouge, LA f d 0.0 0.2 0.4 0.6 0.8 1.0 Cincinnati, OH Present Study, CA k d : Cu 1e+1 1e+2 1e+3 1e+4 1e+5 k d : Zn k d (L/kg) 1.0e+1 1.0e+2 1.0e+3 1.0e+4 1.0e+5 k d : Pb 1.0e+2 1.0e+3 1.0e+4 1.0e+5 1.0e+6 k d : Cd Baton Rouge, LA 1.0e+1 1.0e+2 1.0e+3 1.0e+4 1.0e+5 Cincinnati, OH Present Study, CA k d (L/kg) k d (L/kg) k d (L/kg) Figure 2-11 Comparison of the sample fd and kd of Cu, Zn, Pb, and Cd at the Greater Los Angeles site to other urban and sites

PAGE 49

49 Table 2-1 Hydrologic indices and pave ment residence time data Event Measured Previous Dry Hours Rainfall Duration (hrs) Runoff Duration (min) Rain Depth (mm) Runoff Volume (m3) Event MeanC tc (hrs) mean (hrs) Q50 (L/s) Qmax (L/s) EMIC 20 December 2002 12 February 2003 24 February 2003 15 March 2003 481.47 1273.17 47.92 405.40 8.88 32.00 10.35 23.68 11.53 63.13 19.7 34.27 7.62 86.11 34.54 38.86 610.7 430.6 2954.4 6391.5 0.31 0.54 0.32 0.61 3.02 1.31 1.80 1.94 3.02 3.15 1.92 1.55 9.99 10.56 39.22 52.63 52.63 1067.89 138.31 169.54 All events mean All events median All events SD All events RSD (%) 552.0 443.4 516.6 93.6 18.7 17.0 11.1 59.1 32.2 27.0 22.7 70.6 41.8 36.7 32.6 78.1 2596.8 1782.6 2778.8 107.0 0.44 0.43 0.15 34.5 2.02 1.87 0.72 35.7 2.41 2.47 0.80 33.1 28.1 24.9 21.3 75.8 357.1 153.9 476.4 133.4 WMIC 20 December 2002 12 February 2003 24 February 2003 15 March 2003 481.47 1273.17 47.92 405.40 8.88 32.00 10.35 23.68 11.53 63.13 19.7 34.27 7.62 86.11 34.54 38.86 792.4 370.6 3078.2 8662.5 0.25 0.70 0.24 0.52 2.56 1.27 1.54 1.45 2.56 3.06 2.40 2.10 3.77 10.63 30.73 62.32 75.17 1667.35 138.45 253.29 All events mean All events median All events SD All events RSD (%) 552.0 443.4 516.6 93.6 18.7 17.0 11.1 59.1 32.2 27.0 22.7 70.6 41.8 36.7 32.6 78.1 3225.9 1935.3 3814.6 118.2 0.43 0.38 0.22 52.4 1.71 1.50 0.58 34.1 2.53 2.48 0.40 15.9 26.9 20.7 26.3 97.8 533.6 195.9 759.4 142.3

PAGE 50

50 Table 2-2. Cumulative mass for metals Cu: Particulate (g) Cu: Dissolved (g) Pb: Particulate (g) Pb: Dissolved (g) Ag: Particulate (g) Ag: Dissolved (g) Zn: Particulate (g) Zn: Dissolved (g) WMIC 20 December 2002 26.1 10.4105.14.40.04.0 123.9102.2 12 February 2003 35.5 3.5150.90.40.00.0 46.43.0 24 February 2003 72.9 33.9356.013.40.20.1 262.3419.3 15 March 2003 176.3 91.0931.241.30.70.1 793.8507.8 EMIC 20 December 2002 54.8 10.0154.11.00.03.3 244.817.4 12 February 2003 21.6 4.6107.83.20.00.0 78.217.2 24 February 2003 98.5 32.7370.43.20.50.3 305.6298.6 15 March 2003 193.5 110.6759.828.60.80.0 681.4201.7

PAGE 51

51 Table 2-3. Values of k0 and k1 for Cu, Pb, Zn, and Ag for both catchments WMIC EMIC Parameter k0 or k1 R2 k0 or k1R2 20 December 2002 Cu: Particulate k0=0.978 0.996k1=2.0670.975 Cu: Dissolved k0=0.987 0.999k0=0.9910.999 Pb: Particulate k0=1.017 0.998k1=1.4550.985 Pb: Dissolved k0=0.985 0.995k0=0.9140.945 Zn: Particulate k0=1.017 0.996k1=1.6670.982 Zn: Dissolved k0=0.984 0.999k0=0.8760.927 Ag: Dissolved k0=0.999 0.999k0=1.0110.999 12 February 2003 Cu: Particulate k0=1.019 0.997k1=0.7000.995 Cu: Dissolved k1=1.260 0.996k1=2.2710.998 Pb: Particulate k0=1.022 0.998k1=0.9640.994 Pb: Dissolved k0=0.949 0.995k1=1.4860.998 Zn: Particulate k1=2.037 0.974k1=0.7970.994 Zn: Dissolved k0=0.766 0.907k1=5.3900.995 24 February 2003 Cu: Particulate k1=2.516 0.996k1=2.0850.999 Cu: Dissolved k0=0.996 0.999k0=0.9640.998 Pb: Particulate k1=2.142 0.999k1=1.8040.999 Pb: Dissolved k0=1.015 0.999k0=0.8850.930 Zn: Particulate k1=2.976 0.999k1=2.9530.989 Zn: Dissolved k0=0.958 0.987k0=1.0530.991 Ag: Particulate k1=1.718 0.976k0=1.1990.859 Ag: Dissolved k1=1.392 0.956k1=1.0810.908 15 March 2003 Cu: Particulate k1=1.825 0.997k1=3.6770.998 Cu: Dissolved k0=0.960 0.999k0=0.9190.989 Pb: Particulate k1=1.485 0.995k1=2.8680.998 Pb: Dissolved k0=1.005 0.997k0=0.5890.682 Zn: Particulate k1=1.711 0.998k1=3.0520.999 Zn: Dissolved k0=0.909 0.990k0=0.7630.896 Ag: Particulate k0=0.940 0.970k1=3.2710.972 Ag: Dissolved k1=2.153 0.954 NDND

PAGE 52

52 Table 2-4 Cumulative mass fraction maximum values for various water chemistry parameters TDS (kg) TSS (kg) DOC (kg) TOC (kg) ALK as CaCO3 (kg) Hardness as CaCO3 (kg) WMIC 20 December 2002 132.86 0.039.457.7223.91 68.43 12 February 2003 99.69 337.093.734.728.15 12.89 24 February 2003 1.72 225.4429.8534.1470.27 220.83 15 March 2003 1117.7 1 9.8468.5579.54183.26 364.04 EMIC 20 December 2002 310.39 184.1410.348.700.72 127.57 12 February 2003 35.79 74.254.016.6714.04 26.35 24 February 2003 1002.7 1 708.1034.64124.1578.77 354.95 15 March 2003 2480.4 8 1624.6 397.91113.871098.96 890.83

PAGE 53

53 Table 2-5. Values of k0 and k1 for TDS and TSS for both catchments WMIC EMIC Parameter k0 or k1 R2 k0 or k1R2 20 December 2002 TSS k0=0.990 0.999k1=1.5050.981 TDS k0=0.956 0.988k0=1.0080.995 12 February 2003 TSS k0=1.040 0.993k1=0.6530.995 TDS k0=0.970 0.989k1=2.8270.997 24 February 2003 TSS k1=2.601 1.000k1=2.0980.995 TDS k0=0.978 0.987k0=0.9800.993 15 March 2003 TSS k1=2.228 0.997k1=4.5740.993 TDS k0=0.984 0.992k0=0.9580.998

PAGE 54

54 Table 2-6. Summary of particul ate and dissolved metal EMCs Cu: Particulate [g/L] Cu: Dissolved [g/L] Pb: Particulate [g/L] Pb: Dissolved [g/L] Ag: Particulate [g/L] Ag: Dissolved [g/L] Zn: Particulate [g/L] Zn: Dissolved [g/L] WMIC 20 December 2002 33.0 13.1 132.7 5.6 0.0 5.0 156.3 128.9 12 February 2003 60.0 11.7 287.9 8.4 0.0 0.0 212.6 38.5 24 February 2003 23.7 11.0 107.5 4.4 0.1 0.0 72.2 136.2 15 March 2003 20.3 10.5 107.5 4.8 0.1 0.0 91.6 58.6 All events mean 34.3 11.6 158.9 5.8 0.0 1.3 133.2 90.6 All events median 28.3 11.3 120.1 5.2 0.0 0.0 124.0 93.8 All events SD 18.0 1.1 86.8 1.8 0.0 2.5 64.0 49.3 All events RSD (%) 52.5 9.8 54.6 31.5 115.6 197.4 48.1 54.4 EMIC 20 December 2002 80.5 14.7 226.2 1.4 0.0 4.8 359.5 25.5 12 February 2003 82.5 8.0 350.5 0.9 0.0 0.0 107.8 0.0 24 February 2003 31.6 10.5 118.8 1.0 0.2 0.1 94.3 95.8 15 March 2003 23.0 13.1 90.2 3.4 0.1 0.0 80.9 24.0 All events mean 54.4 11.6 196.4 1.7 0.1 1.2 160.6 36.3 All events median 56.0 11.8 172.5 1.2 0.1 0.0 101.0 24.7 All events SD 31.5 2.9 118.2 1.2 0.1 2.4 133.1 41.3 All events RSD (%) 57.9 25.3 60.2 69.1 122.5 195.6 82.8 113.8

PAGE 55

55 Table 2-7. Summary of EMCs for runoff chemistry parameters TSS [mg/L] TDS [mg/L] TOC [mg/L] DOC [mg/L] ALK [mg/L] Hardness [mg/L] Conductivit y [mho/cm] Turbidity [NTU] pH Redox [mV] WMIC 20 December 2002 77.5 167.7 9.711.930.286.4280.587.3 7.3387.7 12 February 2003 208.1 88.0 11.89.421.631.7103.8392.5 7.8171.6 24 February 2003 73.2 171.6 11.19.722.871.7203.0122.4 7.2233.8 15 March 2003 65.0 65.0 9.27.921.242.0242.4168.9 6.6178.4 All events mean 106.0 123.1 10.59.723.957.9207.4192.8 7.2242.9 All events median 75.4 127.8 10.49.622.256.9222.7145.6 7.2206.1 All events SD 68.3 54.6 1.21.74.225.476.0137.3 0.5100.5 All events RSD (%) 64.4 44.4 11.617.017.643.936.671.2 6.941.4 EMIC 20 December 2002 270.4 455.7 12.815.235.7187.3737.3253.4 7.4379.2 12 February 2003 782.9 231.5 15.59.332.661.2227.3859.0 7.8168.0 24 February 2003 227.1 321.6 39.811.125.30.1422.8326.7 7.1244.4 15 March 2003 192.9 294.5 13.511.640.6105.8357.2416.0 6.8161.7 All events mean 368.3 325.8 20.411.833.588.6436.1463.8 7.3238.3 All events median 248.7 308.0 14.511.434.183.5390.0371.4 7.2206.2 All events SD 278.2 94.5 13.02.56.478.8216.6271.7 0.4101.1 All events RSD (%) 75.5 29.0 63.720.819.188.949.758.6 5.942.4

PAGE 56

56 CHAPTER 3 NUTRIENT PARTITIONING AND TRANSPO RT IN RAI NFALL-RUNOFF FROM AN INSTITUTIONAL SOURCE AREA Introduction The generation, transport and fate of urban source areas nutrients are temporal processes. Nutrient sources include both an thropogenic and biogenic sources. In cases where there is high imperviousness, anthropogenic sources can includ e vehicles from transportation activities and urban practices such as wastewater reuse a nd fertilizer application (Sansalone and Ying, 2008, Novotny and Olem 1994). For example, estimates of particulate matter (P M) fluxes ranged from 26.87 to 94.37 for paved urban areas impacted by ve hicular transportation (Sansalone and Ying, 2008) Studies have demonstrated th at anthropogenic PM results fr om the abrasive interactions of tires, pavement, engine and brake as well as settleable exhaust and maintenance activities (Sansalone, 2001). The biogenic PM results from vegetation sources such as leaf fall, vegetative die-off and landscape maintenance (Mytyk and Delfino, 2004; Sans alone, 2001). As this biogenic material that is of lower specific gravity and in many cases of higher specific surface area, such as for leaves or grass clippings, su ch vegetation can be r eadily transported across hydraulically-efficient paved surfaces during a ra infall-runoff event. In many existing urban land uses, these biogenic material sources are of anthropogenic de sign, such as in a parking lot where there are minimal pervious areas except fo r vegetated or landscaped islands. In such urban designs the relative eleva tion and profile of such island s with respect to the surrounding pavement play a significant role in the transport and fate of biogenic material and reuse loads. For instance, in cases where an island sits relatively close to the immediate surrounding parking lot elevations, effective drainage of pollutants into the islands, instead of transport into a catchment basin, is possible. On the other hand, if the profile of the median is designed to be

PAGE 57

57 significantly higher than the elevatio n of the parking lot, a lot of the biogenic debris that exists can be transported during a storm event. Hydrology coupled with water chemistry has a significant effect on pollutants present in rainfall-runoff. Factors such as the previous number of dry days before a storm event obviously influence the potential quantity of pollutants fo r transport during a storm event (Sheng et al. 2008). Additionally, the intensity of a storm event, as well as the duration influence the transport of rainfall-runoff pollutant constituents (Gnecco et al. 2005). The geometric characteristics of the catchment or watershed plays an integral role in pollutant transport where source area watershed have been shown to generate a more defined and consistent fi rst-flush (mass-limited behavior) as compared to larger and more co mplex watersheds (Sansalone and Cristina 2004). Objectives Comm ensurate with the transport and fate of nutrients and PM at the urban rainfall-runoff interface there are several objectives of this study. The first objective is to compare and contrast the equilibrium concentration of nutrients and PM using non-par ametric analysis on a sample basis for the Gainesville, Florida si te to other urban land uses. This objective is evaluated for dissolved and particulate-bound phases of nitroge n (N) and phosphorus (P). Additionally, the transport behavior (mass-limited or flow-limite d) of the N and P fractions, as well as the sediment, settleable, and suspended PM fractions were evaluated. A third objective further differentiated PM into particle size distributi ons (PSD), modeling PSDs on an event basis. Indices of equilibrium partitioning, sp ecifically the dissolved fraction (fd) and partitioning coefficient (kd) were examined using non-parametric analysis and compared to similar land uses.

PAGE 58

58 Background Saint Johns River Water Manage ment District (SJRWMD) Gainesville, Florida is a part of Alachua County, which is in large part the m ajor contributor of the Orange Creek Basin. This greater basin is approximately 155,400 hectares in size, and the three major water bodies in the Orange Creek Basin are Newnans Lake, Lake Lochloosa, and Orange Lake. The institutional site of interest in this case drains initially to Lake Alice, a lake oncampus at the Un iversity of Florida. The annual rainfall influences the resulting overall runoff that ultimately contributes to the la rger Orange Creek Basin. During the year of 2007, approximately 2600 mm of rainfall was recorded at the institutional site in interest for this study (Wunderground, 2007). From the beginnin g of 1999 to the end of 2003 the Saint Johns River Water Management District (SJRWMD) monitored these th ree waterbodies comprising the Orange Creek Basin. Particular ly, the median total suspende d solids (TSS), total phosphorous, and total nitrogen concentrations were 19 mg/L 0.078 mg/L, and 2.03 mg/L respectively (FDEP, 2001). When comparing these water chemistry para meters to the concentrations listed by FDEP as being the maximum allowable freshwater body concentrations, the results are discouraging (FDEP, 2008). Particularly, th e total phosphorous concentration is of concern in Orange Lake, which is why the Florida Department of Envi ronmental Protection (F DEP) adopted in 2004 a water quality target of 45% reduc tion of its Total Maximum Daily Load (TMDL) (FDEP, 2004). According to FDEP, Orange Lake, which is ap proximately 5440 hectares, showed a general increase in nutrients from 1989 to 1998, and believe that that dissolved oxygen (DO) may be related to the problem (FDEP, 2001). Similarly, in the Orange Creek Basin, FDEP estimates out of the 385,270 acre watershed, and of the 68 miles assessed of lakes, that 50 miles are impaired with nutrients and DO (FDEP, 2001).

PAGE 59

59 Nutrient Concentrations from Paved Surfaces Previous studies have investig ated nutrient partitioning in rainfall-runoff and have shown variant results. For example, an asphalt park ing lot surface in Phoe nix, Arizona found nitrate (NO3 -) and orthophosphate mean concentration to be 26.6 and 0.18 mg/L respectively (Hope et al. 2004). Other studies such as one done in Connecticut showed total Nitrogen (TN) and total Phosphorous (TP) concentrations to be 8.0 a nd 0.244 mg/L respectively (Gilbert and Clausen, 2006). Mytyk and Delfino reported in 2004 that over a 50 year period at 5 of 14 monitored stations there was an increase in the median n itrate concentration in th e Ocklawaha River (Mytyk and Delfino, 2004). More specifically the median nitrate concentration at one of the sites increased from 0.07 mg/L to 0.78 mg/L (Mytyk and Delfino, 2004). Dean et al. (2005) reported nitrate and orthophosphate event mean concentrations of 4.6 and 0.9 mg/L, respectively for a concrete-paved transportation land use in Baton Rouge, LA. When examin ing a study performed on rainfall-runoff in Temple Terrace, FL, the resulting mean TN concentration was 1.14 mg/L, which is significantly lower than the previ ous studies mentioned (Rushton, 2000). Finally, a study done in Ruskin, FL at an agricultural land use found the mean concentrations of organic nitrogen, NO3 -, orthophosphate, and particulate P to be 1.325, 0.385, 0.937, and 0.572 respectively (Rushton, 2002). Mass-Limited vs. Flow-Limited Delivery Over the last century there has been much debate on how pollutant s are tr ansported in rainfall-runoff. Early reference to such transpor t described a first-foul flush of equine fecal matter from urban surfaces (Met calf and Eddy, 1916). While most researchers assume a firstflush (mass-limited) delivery, such behavior repres ents one limit of a tw o limiting classifications of transport from urban surfaces (mass or flow limited) (Sheng et al. 2008). While there are many studies of first flush phenomena for PM and metals (Cristin a and Sansalone 2004,

PAGE 60

60 Sansalone and Buchberger 1997) relatively few studi es have illustrated a first-flush of nutrients. A study that analyzed nutrient transport in Chin a, revealed that both TN and TP experience mass-limited behavior. (Li et al. 2007). Similarl y, a study done in an urban watershed in South Korea determined that TN and TP delivery e xperienced a first flush phenomenon (Lee et al. 2002). Finally, a research study has shown that TN experienced a mass-limited delivery, while TP experienced a flow-lim ited delivery (Geiger, 1987). Site Descriptions The parking lot site in Gainesv ille, Florida, has distinct characteristics in its existing watershed area that result in it being classified as an urban land use. The catchment used in this research, is approximately 450 m2 in size and consists of a pave d parking lot that serves the respective institution. The watershed only catch es a portion of the rainfall-runoff that is generated off the site, because there are several drains that help in catching rainfall-runoff from the parking lot. The basin of interest (Figure 3-1) has one major throughway that vehicles pass on any given day, as well as two parking lanes that are frequently traveled. This catchment has two islands in between traffic lanes that function as partial drainage beca use they are vegetated. In addition, these islands contain trees that both shed nutrients to the parking lot, which are eventually transported during a rainfall-runoff event. The geometric design of these islands in the parking lot transport mo st of the pollutants generated from organic sources, because of the elevation differences in its profile, compared to the remainder of the parking lot. Since the island s are significantly higher than the elevations of the rest of the parking lot, during a storm even t where there are stronge r crosswinds, inevitably pollutants off of these islands are transported to the sampling collection point in the catchment basin.

PAGE 61

61 Additionally, it is important to note that while travelling down the two respective parking lanes from the sample collection point, to the en d of the drainage basin, that only the rainfallrunoff from the centerlin e, along the high-point elevation, are transported to the drainage basin catchment. The rest of rainfall-runoff either being generated directly above or below the centerline are drained to catchments that re side on sides of the parking spaces. The quality of the pavement in the watershe d varies, because there are numerous cracks that exist in the pavement, which cause portions of the stormwater generated during an event to not be transported to the catchment. Additionally, the pavement tends to be extremely rough in some regions, which provides depressional storage for some of the rainfall-runoff. Not only is the pavement quality rough, but tends to have uneven leveling, which cau ses ponds of water to formulate during a storm event that do not imme diately drain to the catchment as expected. Methodology Stormwater Sample Collection In order to ensure quality assurance and quali ty contro l for collecting stormwater samples during each of the four respective storm events, protocol was established with set up of the site, as well as the time intervals and manner that samples were collected. The sample collection point during rainfall-runoff events was set up to collect the stormwater in such a matter to manually retrie ve full cross-sectional samples. This method ensured that accurate representatives of the pollu tant constituents of the stormwater were taken to be analyzed. The rainfall-runoff that was tr ansported to the drainage catchment was first passed through a parshall flume, which enabled fu ll-cross sectional samples to be possible. Traffic Count Besides ensuring proper collection of stormwater sam ples, a determination of the amount of traffic that passes through the catchment on a daily basis was nece ssary. If accurate

PAGE 62

62 estimations of the daily traffic were made, an accurate estimate could be made on the total pollutants that were contributing to the watershe d, because of travelling vehicles. The important assumption made in this process, in order to ach ieve the total contributing PM amounts, was that each vehicle contributed approximately 0.09 mg/m2-vehicle of PM (source?). Because the catchment basin that was being ex amined for pollutants in rainfall-runoff was from an institutional source, typical peak hour tr affic counts were not repr esentative of the total pollutant loads to the watershed. Additionally, the amount of vehicles th at travelled through the catchment basin were significan tly different on a typical wee kday, compared to a typical weekend day. Therefore, traffic counts were performed both on 10 September 2008 and 13 September 2008, to determine daily traffic through the catchment basin. The traffic count on both dates, were done between the hours of 6: 00am to 6:00pm. Vehicles through the watershed were constantly monitored through this time, to be used in future analysis in combination with the amount of previous dry days before a rainfa ll event to determine total pollutant loads. Nutrient Partitioning Once sam ples were collected, proper laboratory analysis was necessary to determine water chemistry parameter concentrations, such as Total Nitrogen (TN), Total Phosphorous (TP), and PM. These parameters in turn could be anal yzed using non-parametric analysis for resulting partitioning and transport during rainfall-runoff events. When examining TN and TP for partiti oning in the resulting rainfall-runoff, NO3 2was considered to be the diss olved portion of TN, while PO4 3was considered to comprise the dissolved portion of TP. The mathematical relationships used to describe the nutrient partitioning of the rainfall-r unoff are described below.

PAGE 63

63 (3-1) (3-2 ) Kd is the partitioning coefficien t (L/kg), P is the particulat e-bound mass of a heavy metal, m is the TSS concentration of the adsorbing so lid material in the aqueous system, D is the dissolved mass of a heavy metal, and fd is the dissolved fraction (Glenn and Sansalone, 2002). EMC Calculations Not only were the p artitioning of the nutrients determined for each respective storm event, but the EMCs were calculated as well. (3-3) In this equation EMC is the flow weighted average concentration for the entire event, M is the total mass of constituent over en tire event duration, V is the to tal volume of flow over entire event duration, c(t) is the time va riable dissolved or particulate bound concentrati on, q(t) is the time variable flow, and t is time. Mass-Limited vs. Flow-Limited Behavior After the E MC analysis, M(V) curves were formulated, and mass-lim ited vs. flow-limited behavior could be determined for each of the storm events through the utilization of the appropriate equations. The first set of equations relate a firs t-order, mass-limited behavior. (3-4) (3-5 ) (3-6) Mt is the cumulative mass delivered, Vt is the cumulative volume, M0 is the constituent mass on the surface at the beginning of the rainfall-runoff event, k1 is the first order coefficient, K is the transport rate coefficient, and M is the instantaneous mass remaining on the watershed surface (Ying and Sansalone, 2008).

PAGE 64

64 The equations below describe flow-limite d events with a zero-order behavior. (3-7) (3-8) (3-9) k0 is the zero-order coefficient, and all othe r parameters are defined as above (Ying and Sansalone, 2008). Particle Size Distribution (PSD) The PSD of PM is of interest in th e rain fall-runoff from each storm event, because it reveals how effective treatment should be designed to remove all or a majority of PM. Once PSD was determined in the laboratory for each sample, PSD could be modeled using a Gamma distribution. The mean particle diameter was take n for each storm event, and could be used with the Gamma distribution to pl ot a representative modele d curve, with corresponding shape, and scale factors. Results Hydrology There are resulting hy drologic characteristics of the four respective storm events (Figure 32, Table 3-1) Of the four respective events, the 05 May 2007 and 02 June 2007 events had lower intensity in rainfall, as well as lower total rainfall volume, while the 15 April 2007 and 29 June 2007 events had higher rainfall intensity, w ith greater runoff volume results. The runoff coefficient at the site ranged from 0.19-0.47, wh ich under initial examina tion, is a relatively low runoff coefficient, considering that the catchment basin almost entirely consists of impervious surfaces, which is directly connected. But when additionally considering the conditions of the site, as described above as having numerous crack s, as well as unlevel paving, these results are not surprising. The runoff coefficient differs with respect to time throughout any given storm

PAGE 65

65 event, for this reason the four respective stor m events the rainfall-runoff coefficients for the catchment were determined, ra nging from 0.13-0.32 (Figure 3-3). Traffic Count As m entioned previously, traffic counts were performed on both 10 September 2008 and 13 September 2008. Results of these traffic counts (Figure 3-4) reveal that this institutional source area watershed has traffic constantly throughout a typical weekday. Additionally, not only is this daily traffic constant, but this traffic has atypical pe aks of other sites because it has its highest volume during the middle of the da y, as well as during the evening traffic rush. M(V) Curves: Particulate Fractions and Particulate N and P Based on the hydrology of the rainfall-runoff even t, the transport of PM can be further analyzed for m ass-limited and flow limited behavior. The particulate fraction, as well as particulate N and P, demonstrated mass-limited be havior for all storm events. The particulate fraction consisting of suspended, settleable, a nd sediment PM all demonstrate strong masslimited behavior (Figure 3-5). There are cumulative totals of mass for each particulate fraction (Table 3-2), and additionally k1 values for these mass-limited events (Table 3-3), as well as their respective R2, coefficient of determination. The partic ulate N and P, similar to the particulate fractions demonstrate mass-limited behavior, but not as strong in some storm events, such as the 15 April 2007 and 05 May 2007 events. M(V) Curves: TDS and Dissolved N and P A m ajority of the storm events additionally have mass-limited behavior for TDS and the dissolved fraction of N and P (Figure 3-6). Ther e are cumulative mass totals for the storm events (Table 3-2), as well as k0 and k1 values (Table 3-4) with their respective R2 values. Although a majority of the storm events for the dissolved fraction of N and P, and TDS display mass-limited

PAGE 66

66 behavior, some flow-limited behavi or exists particularly in NO3 2during the 15 April 2007 and 02 June 2007 storm events. Fractions of PM PM is com prised of three major fractions in cluding suspended, settleable, and sediment. The % that these fractions comprise each sample in a given storm event can be further investigated further (Figure 3-7) Sediment comprises most of the PM, comprising over 90 % in some sample cases. The Suspended Solids Concen tration (SSC) contains all three fractions, or suspended, settleable, and sediment. The SSC can be coordinated with respect to time, in combination with the hydrograph during a given storm to examine when maximum amounts of PM are present. The PSD for each respective storm event is given in Figure 3-8. The modeled curves all have high correlation coefficients, a ll being at least 0.98, and have relatively similar and factors. Nutrient Partitioning There are resulting nutrient partitio ning for both TP and TN (Figure 3-9) The fd range for both TP and TN is relatively lo w, being typically betw een 0.1 to 0.4. This result is expected considering that most of the pol lutants contributing towards the ra infall-runoff can be attributed to debris from trees in the catch ment basin. Similarly, the high kd values demonstrate the tendency of TN and TP being particulate-bound. When comparing these fd and kd results to other urban sites, that are similar in land use to the institutional land use site in this study, it is clear that these values range similarly to other urban sites. The sites used in comparison (Figure 3-10) represent other urban land uses in Florida, as well as one other inst itutional site in Milwaukee, WI.

PAGE 67

67 Total Site Nutrient Loads When analyzing this institutional sites overall im pact on the environment, it is important to consider its total annual nutrient pollutant load s to its surrounding region. Based on the previous dry days for each respective storm event as well as the appropriate assumptions, the expected PM load on the catchment basin can be further analyzed (Figure 3-11) Historically the precipitation to this site during the year 2007 wa s used in this analysis (Figur e 3-11). The site had an annual cumulative rainfall of approximately 2,600 mm. Transl ating this precipitati on into runoff for this site can be done using the ranges of rainfall co efficients already determined, which are 0.130.32 respectively. Using the median value of particulate P, PO4 3-, particulate N, and NO3 2at the site, a median load can be estimated using the high an d low range of these nutrients load. There are estimated loads throughout 2007 for the nutrients mentioned above, as well as PM and TDS (Figure 3-11). Note the resulting estimated anthropogenic cont ributions of PM, based on the previous dry days for the four respective st orm events (Figure 3-11). Resu lts were calculated using the information gained from traffic counts performe d. Because the 5 May 2007 event had the most previous dry days it had the largest estim ated contribution of approximately 652.32 mg/m2 of PM. This estimated value corr esponds with the measured PM concentrations for the 05 May 2007 event, which were significantly hi gher than the other storm events. EMC of Water Chemistry Parameters The EMC of relevant water chem istry paramete rs are summarized (Table 3-5). However, it is important to compare these EMCs to those of other urban sites that are similar in land use, in this case being institutional. A comparison is pr ovided of the site in Gainesville, FL to other similar land uses in Florida, as well as another in stitutional site in Milwaukee, WI (Figure 3-12). Generally the concentration of TP, TN and TSS are higher at this part icular site than other sites.

PAGE 68

68 This can be attributed to the proximity of organisms generating PM and nutrients close to the catchment in the watershed. Implications/Discussion There ar e significant implications of this re search associated with the partitioning and transport of nutrients in rainfall-runoff. Regulators with the public sector, such as in this case the SJRWMD, need to carefully mon itor surrounding waterbodies to en sure that urbanization is not increasing the overall nutrient concentrations pres ent in these respective waterbodies. Increased levels of TN and TP concentrations adversely affect a freshwater body environment. Increased levels of these types of nutrients result in eutrophication, and ultimately result in an overproduction of algae organisms, that decreas e the dissolved oxygen concentrations available for larger organisms such as fish. Additiona lly, when smaller PM is still suspended in waterbodies, it can clog the gills of fish and cause more difficulty in breathing, which is vital to a sustainable life. Besides the implications that exist for regul ating pollutants in rainfall-runoff for the preservation of organisms in lo cal waterbodies, it is important for regulators to consider the effectiveness of the varying desi gn alternatives that exist for tr eatment of stormwater. Options such as basins, hydrodynamic separators, fine solid separators, and filters such all be considered for effective rainfall-runoff treatment. When maki ng the final decision on what treatment is best, removal efficiencies of the pollutant constituen ts such be considered, as well as the cost effectiveness of the chosen desi gn alternative for implementation. Conclusions In conclusion, even a smaller urban wate rshed can contribute significant am ounts of nutrient pollutant loads to its surrounding region. Th is is a reality, especial ly in situations where the geometric design of impervious ar eas such as in parking lots are raised in such a way that all

PAGE 69

69 pollutants are transported in rainfall-runoff, instead of settling into the vegetated soils that exist in the medians. Redesign of this particular site in Gainesville, Florida c ould result in significant reductions of pollutants transported in rainfall-runo ff, especially if the islands in the parking lot were lowered in elevation. Additionally, swale drainage could be initiated instead of the curb and gutter currently present at the site. Other urban land uses should consider carefully how the geometric design they choose is affecting its su rrounding environment, particularly the pollutants that can result in rainfall-runoff from a site. Nutrients, both TN and TP, experience partiti oning in rainfall-runoff that is predominantly in the particulate-bound st ate, with a mean fd value of 0.21 and 0.45 for N and P respectively. This reality can be taken into consideration in future design, to ensure proper removal of pollutant constituents. Additionally, a majority of the PM present in th e rainfall-runoff exists in the sediment form of the three fractions, in over 90% in some cases. This similarly should be taken into consideration when designing effective treatm ent design to promote full removal of PM. Although sediment exists as the majority of PM, suspended and settleable solids should still be removed in entirety through correct BMP. Because a majority of PM exhibit a first-flush phenomena, particulate P, particulate N, and SSC, with mean k1 values of 5.67, 5.97, and 4.58 respectively, can be readily removed during the first half of a rainfall-runoff event with the implementation of the best treatment strategies. Similarly the dissolved fraction of P and N, as well as TDS, experienced a mass-limited transport, but not as strong of a fi rst-flush as the PM, with median k1 values of 3.84, 3.26, and 7.57 respectively. Therefore, caref ul attention should be paid to the entire spectrum of delivery of pollutant constituents to accomplish full removal.

PAGE 70

70 Finally, these results prove that the PSD of rainfall-runoff produced from storm events vary in particle diameter from less than1 m to approximately100 m. For this reason, when designing treatment of the PM, it is important to recognize these inhe rent differences in particle diameter to accurately colle ct all forms of solids.

PAGE 71

71 Figure 3-1. Plan view of catchment

PAGE 72

72 0 50 100 150 0 1000 2000 3000 4000 5000 6000 Rainfall (mm/hr) 0 50 100 150 2 June 2007 0 5 10 15 20 25 0 200 400 600 800 1000 Rainfall (mm/hr) 0 5 10 15 5 May 2007 Runoff (L/min) 0 20 40 60 80 100 Cumulative Runoff (L) 0 200 400 600 800 1000 0 10 20 30 40 29 June 2007 Elapsed Time (minutes) 020406080100120Runoff (L/min) 0 200 400 600 800 1000 Cumulative Runoff (L) 0 1000 2000 3000 4000 5000 6000 0 50 100 150 15 April 2007Runoff (L/min) Cumulative Runoff (L) Cumulative Runoff (L) Runoff (L/min) Rainfall (mm/hr) Rainfall (mm/hr) Figure 3-2. Event-based hydrology

PAGE 73

73 t/tmax 0.00.20.40.60.81.0 Rainfall-Runoff Coefficient, C 0.0 0.1 0.2 0.3 0.4 0.5 5 May 2007 2 June 2007 15 April 2007 29 June 2007 tmax = 55 min tmax = 116 min tmax = 96 min tmax = 57 min Figure 3-3. Event-based rainfall-runoff coefficient

PAGE 74

74 Vehicles 0 10 20 30 40 10 September 2008 ( Vehicles 0 10 20 30 40 13 September 2008 ( = 84) Time (hours) 024681012 Vehicles 0 10 20 30 40 24 October 2008 ( = 379) Figure 3-4. Traffic count for catchment ba sin during a typical weekday and weekend.

PAGE 75

75 Particulate Phosphorous M/M max 0.0 0.2 0.4 0.6 0.8 1.0 15 April 2007 05 May 2007 02 June 2007 29 June 2007 Particulate Nitrogen Suspended PM M/M max 0.0 0.2 0.4 0.6 0.8 1.0 Settleable PM Sediment PM V/V max 0.00.20.40.60.81.0 M/M max 0.0 0.2 0.4 0.6 0.8 1.0 SSC V/V max 0.00.20.40.60.81.0 1 1 _______ Figure 3-5. Mass-limited behavior of partic ulate fractions, and particulate N and P

PAGE 76

76 M/M max 0.0 0.2 0.4 0.6 0.8 1.0 15 April 2007 05 May 2007 02 June 2007 29 June 2007 NO3 M/M max 0.0 0.2 0.4 0.6 0.8 1.0 PO4 3TDS V/V max 0.00.20.40.60.81.0 M/M max 0.0 0.2 0.4 0.6 0.8 1.0 1 1 _______ Figure 3-6. Mass-limited and flow-limited behavi or of dissolved Nitrogen, Phosphorous, and Total Dissolved Solids

PAGE 77

77 15 April 2007 Mass Fraction, % 0 20 40 60 80 100 C/C max 0.0 0.2 0.4 0.6 0.8 1.0 t/t max 0.00.20.40.60.81.0 C/C max 0.0 0.2 0.4 0.6 0.8 1.0 02 June 2007 Mass Fraction, % 0 20 40 60 80 100 Suspended PM Settleable PM Sediment PM 29 June 2007 05 May 2007 t/t max 0.00.20.40.60.81.0 Hydrograph SSC 1 2 3 4 5 6 7 1 2 3 4 5 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 9 10111213 Cmax = 278.3 mg/L Cmax = 7797.0 mg/L Cmax = 88.1 mg/L Cmax = 922.7 mg/L Figure 3-7. Fraction of suspended, settleable, and sediment vs. the maximum SSC concentration throughout the hydrograph

PAGE 78

78 15 April 2007 % Finer By Mass 0 20 40 60 80 100 05 May 2007 Mean (measured) Modeled 29 June 2007 Particle Diameter ( m) 0.1 1 10 100 1000 n = 5 = 0.730 = 41.20 R2 = 0.99 n = 9 = 0.606 41.88 R2 = 0.98 n = 13 = 0.989 38.34 R2 = 0.99 02 June 2007 Particle Diameter ( m) 0.1 1 10 100 1000% Finer By Mass 0 20 40 60 80 100 n = 7 = 0.836 27.66 R2 = 0.99 PSD bounds Figure 3-8. Event-based particle size distribution of PM

PAGE 79

79 TP f d 0.0 0.2 0.4 0.6 0.8 1.0 05 May 2007 02 June 2007 29 June 2007 TN TP 15 April 2007 kd (L/kg) 1e+1 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 TN 15 April 2007 05 May 2007 02 June 2007 29 June 2007 Figure 3-9. Dissolved fraction and equilibrium partitioning coefficient for Phosphorous and Nitrogen

PAGE 80

80 TP f d 0.0 0.2 0.4 0.6 0.8 1.0 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 f d k d TN Gainesville, FL f d 0.0 0.2 0.4 0.6 0.8 1.0 k d (L/kg) 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 1e+8 Temple Terrace, FL Tampa, FL Ruskin, FL SWFWMD, FL Milwaukee, WI k d (L/kg) Figure 3-10. Comparison of dissolved fraction and partitioning coefficient for other similar sites that represent parking of pavement

PAGE 81

81 Cumulative N Load (g/m 2 ) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Range Part N Load Range NO3 Load Median: Part N Median: NO3 Year: 2007 Cumulative Runoff (mm/m2) 0 100 200 300 400 500 600 Range of Runoff Rainfall (mm) 0 20 40 60 80 100 120 140 160 180 200 Cumulative Rainfall (mm) 1400 1600 1800 2000 2200 2400 2600 2800 Year: 2007 Cumulative P Load (g/m 2 ) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Range Part P Load Range PO4 3Load Median: Part P Median: PO4 3Year: 2007February April June August October December Cumulative PM Load (g/m 2 ) 0 20 40 60 80 Range PM Load Range TDS Load Median: PM Median: TDS C = 0.47 C = 0.19 Year: 2007 Year: 2007January March May July September November February April June August October December January March May July September November Cumulative TN Load (g/m 2 ) 0.0 0.1 0.2 0.3 0.4 0.5 Range Atmospheric NO3 2Load Median: Atmospheric NO3 215 April 07: 198.72 mg/m205 May 07: 652.32 mg/m202 June 07: 219.78 mg/m229 June 07: 136.44 mg/m2Anthropogenic PM Figure 3-11. Total annual nutrient loads of catch ment basin for total phosphorous, nitrogen, and PM based on the high and low runoff coefficients

PAGE 82

82 Nitrogen [mg/L] 0.001 0.01 0.1 1 10 100 TN: Particulate NO2 + NO 3 Phosphorous [mg/L] 0.001 0.01 0.1 1 10 100 TP: Particulate PO 4 3Gainesville, FL TSS [mg/L] 1e+0 1e+1 1e+2 1e+3 1e+4 1e+5 Temple Terrace, FL Tampa, FL Ruskin, FL SWFWMD, FL Milwaukee, WI Figure 3-12. Comparison of partic ulate and dissolved nitrogen and phosphorous to other similar sites that represent parking of pavement

PAGE 83

83 Table 3-1. Summary of hydrology Event Measured Previous Dry Time (days) Rainfall Duration (min) Runoff Duration (min) Rain Depth (mm) Runoff Volume (L) C tc (min) tmean (min) Q50 (L/min) Qmax (L/min) 15 April 2007 5 May 2007 2 June 2007 29 June 2007 4.02 11.81 8.00 3.95 30 40 75 101 57 55 96 116 22.9 2.0 2.3 1.4 816.2 751.1 474.5 5264.7 0.19 0.33 0.21 0.47 27.0 25.0 32.0 19.0 27.0 7.0 16.1 6.9 12.4 2.2 3.0 0.2 109.1 84.2 21.9 960.0 All events mean All events median All events SD All events RSD (%) 6.95 6.01 3.76 54.07 62 58 33 53 81 77 30 37 9.7 6.9 9.8 101.9 1826.6 783.6 2296.9 125.7 0.30 0.27 0.13 42.0 25.8 26.0 5.4 20.9 14.2 11.5 9.6 67.2 4.4 2.6 5.4 122.6 293.8 96.7 445.7 151.7 C: Runoff Coefficient tc: Initial lag time (min) tmean: Mean lag time (min) Q50: Median flow rate during storm event (L/min) Qmax: Maximum flow rate during storm event (L/min)

PAGE 84

84 Table 3-2. Cumulative total of phosphorous nitrogen, and particulate matter Storm Event P: Particulate [mg/L] PO4 3[mg/L] N: Particulate [mg/L] NO3 [mg/L] Suspended PM [mg/L] Settleable PM [mg/L] Sediment PM [mg/L] SSC [mg/L] 15 April 2007 15.8 7.712.123.2149.2129.4 177.9456.5 5 May 2007 52.7 25.318.026.5852.5714.8 12267.213834.5 2 June 2007 2.0 5.612.81.626.714.0 108.1148.7 29 June 2007 23.8 6.717.34.2609.2530.9 2905.34045.4

PAGE 85

85 Table 3-3. Values of k0 and k1 for particulate-bound N and P, and PM Storm Event k0 or k1 R2 Particulate P 15 April 2007 k1= 3.26 0.99 5 May 2007 k1= 1.80 0.97 2 June 2007 k1= 7.82 0.98 29 June 2007 k1= 9.78 0.96 Particulate N 15 April 2007 k1= 1.86 0.95 5 May 2007 k1= 2.93 0.98 2 June 2007 k1= 4.63 0.99 29 June 2007 k1= 14.48 0.75 Suspended PM 15 April 2007 k1= 2.65 0.87 5 May 2007 k1= 1.95 0.97 2 June 2007 k1= 2.32 0.99 29 June 2007 k1= 7.26 0.92 Settleable PM 15 April 2007 k1= 8.79 0.96 5 May 2007 k1= 1.61 0.94 2 June 2007 k1= 3.56 0.99 29 June 2007 k1= 4.01 0.99 Sediment PM 15 April 2007 k1= 9.30 0.94 5 May 2007 k1= 2.85 0.91 2 June 2007 k1= 5.23 0.77 29 June 2007 k1= 5.15 0.98 SSC 15 April 2007 k1= 5.73 0.97 5 May 2007 k1= 2.74 0.92 2 June 2007 k1= 4.65 0.85 29 June 2007 k1=5.19 0.99

PAGE 86

86 Table 3-4. Values of k0 a nd k1 for PO43-, NO3-, and TDS Storm Event k0 or k1 R2 PO4 315 April 2007 k1= 2.59 0.96 5 May 2007 k1= 2.84 0.97 2 June 2007 k1= 3.84 0.99 29 June 2007 k1= 20.55 0.60 NO3 15 April 2007 k0= 0.92 0.99 5 May 2007 k1= 3.26 0.97 2 June 2007 k1= 2.12 0.90 29 June 2007 k1= 23.87 0.75 TDS 15 April 2007 k0= 0.89 0.98 5 May 2007 k1=3.40 0.91 2 June 2007 k1= 7.57 0.98 29 June 2007 k1= 16.63 0.74

PAGE 87

87 Table 3-5. EMC of particulate and dissolved phosphorous, nitrogen, and particulate matter Storm Event P: Particulate [mg/L] PO4 3[mg/L] N: Particulate [mg/L] NO3 [mg/L] Suspended PM [mg/L] Settleable PM [mg/L] Sediment PM [mg/L] SSC [mg/L] 15 April 2007 3.5 1.6 11.5 4.7 149.2 129.4 177.9 79.5 5 May 2007 8.4 3.4 2.5 3.5 115.0 124.3 2045.0 2284.4 2 June 2007 0.6 0.8 1.7 0.2 4.3 2.0 9.2 15.4 29 June 2007 2.0 0.3 1.1 0.2 51.7 71.7 385.6 86.4 All events mean 3.6 1.5 4.2 2.1 80.1 81.8 654.4 616.4 All events median 2.7 1.2 2.1 1.9 83.4 98.0 281.7 82.9 All events SD 3.4 1.3 4.9 2.3 64.7 59.3 939.8 1112.4 All events RSD 93.8 86.8 116.9 107.4 80.8 72.4 143.6 180.5

PAGE 88

88 CHAPTER 4 CONCLUSIONS Pollutant Concentrations It is im portant to compare the resulting con centrations of these rainfall-runoff constituents to other reported results. When comparing the concentrations of the metals present in stormwater from the industrial s ite to those present in the Nati onal Stormwater Quality Database (NSQD), they are within similar ranges in some cases. For instance, Zinc (Zn) had a mean site concentration of 143.7 g/L, which is comparab le to the NSQD median concentration of 200 g/L respectively (Pitt et al. 2004). In contrast, the mean concentration of Lead (Pb) at the site of interest, being 181.4 g/L, was higher than th e median concentration of the NSQD, compared to 26 g/L respectively. Pollutant Loads W hen examining the overall pollutant load s consisting of particulate and dissolved nitrogen (N) and phosphorus (P), as well as par ticulate matter (PM) and total dissolved solids (TDS), it is evident that a parking lot source ar ea contributes a significa nt amount of pollutants annually. For instance, the expected annual partic ulate and dissolved P load is approximately 0.8 g/m2 and 0.4 g/m2 respectively. The N loads experiences transport at a similar magnitude, while PM and TDS is expected to deliver 40 g/m2 and 25 g/m2 annually respectively. Partitioning of Pollutants The partitio ning of pollutants exhibited by the two land uses studied in this research, (industrial and parking lot), have de monstrated results comparable to other studies. Particularly with the industrial land use, metals displayed a dissolved fraction (fd) similar to research studies in the past. More specifi cally, Pb being mainly part iculate-bound, having a mean fd value of 0.036 at the industrial site, is expected b ecause of known bonding preference theories of

PAGE 89

89 hydrolysis (McBride, 1994; Baes, and Messmer 1976; Burgess, 1978; Sansalone and Glenn, 2007). The other metals, being C u, Cd, and Zn, can range in fd in greater magnitudes when compared to other sites such as at other urban land uses in Cincinnati, OH and Baton Rouge, LA (Dean et al. 2005, and Sansalone and Buchberger, 1997). When examining the parking lot land use, a majority of the nutrients was exhibited in the particulate bound state, with respect to phosphorus and nitrogen concentrations, because of the biogenic sources of pollutants in close proximity to the catchment. Transport of Pollutants The transpo rt of pollutants at both of the respective land uses had va rying results in masslimited and flow-limited delivery of pollutants. Th e size of the watershed has been known in the past to effect the transport of pollutants, with larger watersheds typically not exhibiting a first flush phenomenon of pollutants (San salone and Cristina, 2004). In contrast, small catchments exhibit more of a first flush (Sansalone and Cristin a, 2004). This case was true in the watersheds studied in this research because the larger watershed of approximately 73 ha from the industrial land use displayed mainly flow-lim ited behavior from the dissolved portion of its metal pollutant constituents. Contrastly, at th e institutional land use, where there was a significantly smaller watershed of approximately 450 m2, more mass-limited behavior was exhibited for both the particulate and dissolved portion of nutrients. Although the industrial land use did exhibit some similarity to the institutional land use, result s indicated mass-limited delivery in some storm events for particulate bound metals. Industrial Water Reuse Instead of the storm water generated onsite at the industrial so urce area watershed in greater Los Angeles being wasted by being delivered to the Dominguez Channel, consideration should be given to implementing a treatment system th at would purify the rainfall-runoff to a level

PAGE 90

90 acceptable for an industrial water reuse applicatio n. A treatment system for the rainfall-runoff would provide multiple benefits to the surroundi ng community. For example, local aquatic life would no longer be adversely affected by the rainfall-runoff entering their environment. Additionally, once the stormwater is treated, it can be used as a re source to industries in the local area that need evaporative cooling water, bo iler feedwater, process water, and water for landscaping and maintenance of i ndustrial grounds (Tchobanoglous et al. 2003). The types of industries that typically have th ese needs include steel, chemical and allied products, paper and allied products, and petroleum refining i ndustries (Tchobanoglous et al. 2003). There have been numerous examples both in the United States (U S) and throughout the world where reuse water has been applied for indu strial use applications Probably one of the most widely known examples of a town wh ere industrial ecology practices have been implemented is in Kalundborg, Denmark. This town has four main industries including a 1,500megawatt coal-fired plant, a la rge oil refinery, a maker of ph armaceuticals and enzymes called Novo Nordisk, and a plasterboard manufacturer cal led Gyproc (Ehrenfeld and Gertler, 1997). There are eleven linkages that exist in Kalundbor g that make use of waste streams and energy resources (Ehrenfeld and Gertler, 1997). One of the significant waste to resource streams in Kalundborg is in the 700,000 cubic me ters per year of resuse water that an oil refinery has pumped to the coal fired power plant (Ehrenfeld and Gertler, 1997). The received reuse water at the power plant is purified and used as boi ler feedwater (Ehrenfeld and Gertler, 1997). Additionally, the power plant gi ves treated wastewater, approx imately 200,000 cubic meters per year, to the oil refinery for use in cleani ng purposes (Ehrenfeld and Gertler, 1997). Another example of reuse appl ications is at the Palo Verd e Nuclear Generating Station in Arizona (Tchobanoglous et al. 2003 ). Wastewater effluent fr om Tolleson and Phoenix is

PAGE 91

91 pumped 38 miles to this station for cooling tower operations (Tchobanoglous et al. 2003). Before this reuse water is applied, it is treated using biological nitrific ation, lime and soda ash addition for softening and phosphorus removal, f iltration, and chlorinati on (Tchobanoglous et al. 2003). This treatment effectively prevents sca ling and reduces corrosion in the cooling tower systems (Tchobanoglous et al. 2003). In 1995 it was estimated that 82 % of industria l water use was from surface water, while 18 % was attributed to groundwater, and reclaime d water was less than one % (Tchobanoglous et al. 2003). Part of the reason for the slow applicat ion of reuse water in the industry is due to the concern of water quality problems th at result in cooling tower syst ems such as scaling, metallic corrosion, biological growth, and fouling in heat exchanger and condensers (Tchobanoglous et al. 2003). Scaling occurs in cooling towers when there is formation of hard deposits, that reduce the efficiency of heat exchange (Asano et al. 2007). These hard deposits are primarily in the form of calcium carbonate, calcium sulfate, and cal cium phosphate, but magnesium carbonate and phosphate can pose as a threat as well (Asano et al. 2007). Scaling can be avoided by removing phosphate with precipitation, or with methods su ch as ion exchange (Asano et al. 2007). However, ion exchange is more expensive is more expensive than th e method of precipitation, and therefore is not the prefe rred option (Asano et al. 2007). Besides scaling posing as a problem in cooling tower systems, metallic corrosion is a cause of concern as well. Corrosion occurs when th ere is an electrical potential between dissimilar metal surfaces (Tchobanoglous et al. 2003). A ma jor water quality parameter that contributes towards metallic corrosion is TDS, because it increases the electrical conductivity of the solution. Besides TDS, dissolved oxygen and cer tain metals such as manganese, iron, and

PAGE 92

92 aluminum, have high oxidation pote ntials, and therefore promote co rrosion (Tchobanoglous et al. 2003). Techniques of preventing metallic corrosion include adding chemical corrosion inhibitors (Tchobanoglous et al. 2003). Not only are scaling and metallic corrosion water quality problems in cooling tower systems, but biological growth can be a problem associated with reuse water as well. The primary water quality problems attributed to biol ogical growth are nutrients such as nitrogen (N) and phosphorus (P). Biological growth results in the growth of microorganisms that attach and deposit on heat-exchanger surfaces, and eventually l eads to an inhabitance of heat transfer and water flow (Asano et al. 2007). Biological growth can be prevented adding biocides in the internal chemical treatment process (Asano et al. 2007). Finally, fouling presents con cern in using reuse water in cooling tower applications. Fouling involves the attachment and growth of deposits in cooling towe r recirculation systems (Tchobanoglous et al. 2003). Specifically thes e deposits consist of biological growths, suspended solids, silt, corrosion products, and in organic scales (Tchobano glous et al. 2003). Fouling leads to inhibition of transfer in heat exchangers, and can be prevented by adding chemical dispersants that prevent particle aggregation (Tchobanoglous et al. 2003). Additionally, methods of preventi ng fouling include simply using the chemical coagulation and filtration processes that are required for phosphor us removal (Tchobanoglous et al. 2003). If the rainfall-runoff generated onsite at the industrial source area watershed in greater Los Angeles was used for cooling tower applications, the maximum concentrations of water quality parameters acceptable for application are listed in Table 4-1 Irrigation Reuse Water The rainfall-runoff generated onsite at the inst itutional land u se, with proper initiation of treatment, could be suitable for reuse water landscape irrigation applicat ion. Caution must be

PAGE 93

93 followed in applying this reuse water to ensure that problems associated with salinity, specific ion toxicity, and water infiltration rate do not occur with the local vegetation. Salinity is probably the most determinant f actor in deciding whethe r effluent water is acceptable for reuse in an irrigation application. Salinity is closely relate d to the presence of TDS, which can be correlated to electrical conductivity (Asano et al. 2007). Salinity can negatively affect plant growth because there can possible osmotic effects, specific ion toxicity, and soil particle dispersion (Asano et al. 2007). Furt her, increased soil salinity in the roots of plants can cause expenditure of energy on adjusting the salt concentration within tissue, instead of expending the energy on plant growth (Asano et al. 2007). Specific ion toxicity occurs when there is a decline in plant grow th due to excessive concentrations of specific ions, such as sodium chloride, and boron (Tchobanoglous et al. 2003). Boron presence is usually due to household detergen ts or part of the discharges from industrial plants (Tchobanoglous et al. 2003). Presence of chloride and sodium can be attributed to domestic usage, with water soften ers (Tchobanoglous et al. 2003). Besides specific ion toxicity, a concern of the application of reuse water for irrigation use is with the water infiltration rate. High sodium in reuse water can led to deterioration of the physical condition of a soil, more specifically the formation of crusts, waterclogging, and reduced soil hydraulic conduc tivity (Asano et al. 2007). Problems with regions where there is already soil with permeability problems can be resolved by excavating and rearranging the relevant land (Asa no et al. 2007). Other problems that can occur with the a pplication of rainfall-runoff reuse water in irrigation can be associated with nutrients. Although nutrients are fundamental as a fertilizer, in excessive amounts nutrients such as N and P can cause excessive vegetative growth, delayed or

PAGE 94

94 uneven maturity, or reduced crop quality (Tchob anoglous et al. 2003). Solutions toward reuse water with high nitrogen levels is to blend the reuse water with other water supplies or to apply seasonal denitrification (Tchobanoglous et al. 20 03). Not only are nutrients of concern, but clogging problems can sometimes occur with reuse water, as well as biol ogical growth in the sprinkler head or supply line, which result s in plugging (Tchobanoglous et al. 2003). In order to avoid these issu es that can occur with the a pplication of reuse water in irrigation, maximum concentrations should not be exceeded for relevant water quality parameters. Table 4-2 yields some appropria te water quality standards based on a water application rate of 1.25 m/yr. Global Conclusions In conclusion, depending on the land use, quality and quantity of rainfall-runoff being generated at a particular site, c onsideration should be given to reuse water applications. Rainfallrunoff should be treated adequately to meet wa ter chemistry treatment standards needed for irrigation and cooling tower use a pplications. Reuse water applica tion will not only result in less local impact on surface water bodies with less eutrophication occurr ing with excessive nutrients, and less negative aquatic life impacts associated with metals, but will result in a reduced demand on potable water supply that is a resource that is quickly beco ming more expensive to obtain, treat, and distribute on a regular basis. O ffsetting potable water supply demands in local communities will result in a more sustainable practic e that can make this resource last longer for future generations.

PAGE 95

95 Table 4-1. Water quality requirements for us e in steam generation and cooling in heat exchangers. Boiler Feedwater Boiler Feedwater Boiler Feedwater Cooling Water Cooling Water Cooling Water Cooling Water Characteristic Low Pressure Intermediate Pressure High Pressure Once-through Fresh Once-through Brackish Once-through Fresh Once-through Brackish Silica 30 100.7502550 25 Aluminum 5 0.10.01N/AN/A0.1 0.1 Iron 1 0.30.05N/AN/A0.5 0.5 Manganese 0.3 0.10.01N/AN/A0.5 0.5 Calcium N/A 0.40.01N/AN/A0.5 0.02 Magnesium N/A 0.250.01N/AN/AN/A N/A Ammonia 0.1 0.10.1N/AN/AN/A N/A Bicarbonate 170 1204860014024 140 Sulfate N/A N/AN/A6802700200 2700 Chloride N/A N/AN/A60019000500 19000 Dissolved solids (TDS) 700 500200100035000500 35000 Copper 0.5 0.050.05N/AN/AN/A N/A Zinc 0.010.01N/AN/AN/A N/A Hardness 350 10040500115350 115 Alkalinity 350 10040500115350 115 pH 7.0-10.0 8.2-10.08.2-9.05.0-8.36.0-8.3 Chemical Oxygen Demand 5 51757575 75 Suspended Solids 10 50.550002500100 100 Source: NAS (1972). Note: Units are mg/L. Brac kish Water = dissolved solids more than 1000 mg/L. N/A = Not Applicable (Never has been a problem at concentrations encountered)

PAGE 96

96 Table 4-2. Recommended maximum concentrations of water quality parameters in irrigation water Element Concentration (mg/L) Aluminum 5.0 Arsenic 0.10 Cadmium 0.010 Cobalt 0.050 Fluoride 1.0 Lithium 2.5 Nickel 0.20 Lead 5.00 Zinc 2.0 Source: Ayers and Wescot (1985)

PAGE 97

97 LIST OF REFERENCES Am erican Public Health Association (APHA). ( 1998). Standard methods for the examination of water and wastewater, 20th Edition, A. D. Eaton, L. S. Clesceri, and A. E. Greenberg, eds., American Public Health Association, Am erican Water Works Association, and Water Environment Federation, Washington, D.C. Anderson, B.C., Bell, T., Hodson, P., Marsalek, J., and Watt, W.E. (2004). Accumulation of Trace Metals in Freshwater Invertebrates in Stormwater Management Facilities. Water Quality Research Journal Canada : 39 (4) pp. 362-373. Arnold, C., and Gibbons, J. (1996). Impervious surface coverage: the emergence of a key environmental indicator. Journal of the American Planning Association ,62(2). pp. 243258. Asaf, L. (2004). Controls on the chemical and isotopic compositions of urban stormwater in a semiarid zone. Journal of Hydrology 294, pp. 270-293. Asano, T., Burton, F.L., Leverenz, H.L., Tsuc hihashi, R., and Tchobanoglous, G. (2007). Water Reuse Issues, Technologies, and Applications. Metcalf and Eddy, Inc. McGraw Hill, Inc., New York. Ayers, R.S. and D.W. Wescot (1985). Wate r Quality for Agriculture, FAO Irrigation and Drainage Paper 29, Rev. 1, Food and Agriculture Organization of the United Nations, Rome, Italy. Ayres, R. U. (1992). Toxic metals: materials cy cle optimization. Proceedings of the National Academy of Sciences of the United States of America, 89(3), pp. 815-820. Baes, C.F. and Messmer, R.E. The Hydrolysis of Cations (1976) Wiley-Interscience, New York. Bedient, P. B., and Huber, W. C. (1992). Hydrology and Floodplain Analysis Addison Wesley Publishing, Massachusetts. Bertrand-Krajewski, Jean-Luc et al. (1998). Distribution of pollutant mass vs volume in stormwater discharges and the first flush phenomenon. Water, Air, and Research Journal 32(8), pp. 2341-2356. Bitton, G. (2005). Wastewater Microbiology John Wiley & Sons, Inc. Hoboken, New Jersey. pg. 88. Bos, M.G. (1989). Discharge Measurement Structures International Institute for Land Reclamation and Improvement. Wage ningen, The Netherlands. pp. 32-33. Brabec, E., Schulte, S., and Richards, P.L. ( 2002). Impervious surfaces and water quality: a review of current literatu re and its implications for watershed planning. Journal of Planning Literature, 16(4), pp. 499-514.

PAGE 98

98 Brezonik, P.L., E.C. Blancher, V.B. Myers, C.L.H ilty, M.K. Leslie, C.R. Kr atzer, G.D. Marbury, B.R. Snyder, T. L. Crisman & J. J. Messe r (1979). Factors aff ecting primary production in Lake Okeechobee, Florida. Rept. N o. 077901, Dept. Env. Eng. Sci., Univ. Florida, Gainesville, FL: 294 pp. Burgess, J. Metal Ions in Solution (1978). Ellis Horwood, Chichester. pp. 264-267. Carpenter, S.R., Caraco, N.F., Correll, D.L., Howarth, R.W., Sharpley, A.N., and Smith, V.H. (1998). Nonpoint Pollution of Surface Wate rs with Phosphorus and Nitrogen. Ecological Applications : 8(3). pp. 559-568. Christensen, E. R. and Soonthornnonda, P. (2005). Milwaukee Metropolitan Sewerage District (MMSD) stormwater monitoring program final report 2000-2004. University of Wisconsin Milwaukee. pp. 99, 107, 115, 147, 151. Couillard, Y., Campbell, P.G.C., Tessier, A., Pelle rin-Massicotte, J., and Auclair, J.C. (1995). Field transplantation of a freshwater bivalve, Pyganodon gra ndis, across a metal contamination gradient. I. Temporal change s in metallothionein and metal (Cd, Cu, and Cd) concentrations in soft tissues. Canadian Journal of fisheries and aquatic sciences : 52 (4). pp. 690-702. County of Los Angeles Department of Public Works (2000). Seasonal summary of hydrologic data for monitored stations. http://ladpw.org/wmd/NPDES/94 00_tbl_list.cfm Accessed 18 May 2008. Davis, A. P., Shokouhian, M., and Ni, S. (2001). Loading estimates of lead, copper, cadmium, and zinc in urban runoff from specific sources. Chemosphere, 44, pp. 997-1009. Dean, C. M., Sansalone, J. J., Cartledge, F. K., and Pardue, J. H. (2005). Influence of hydrology on rainfall-runoff metal element speciation. Journal of Environmental Engineering, 131(4), pp. 632-642. Ehrenfeld, J. and Gertler, N. (1997) Industrial Ecology in Practice. Journal of Industrial Ecology 1(1). pp. 67-79. Florida Department of Environmental Prot ection (FDEP) Division of Water Resource Management. (2001). Central District, Group 1 Basin, Basin Status Report Ocklawaha. Florida Department of Environmental Prot ection (FDEP). (2008). 62302.530: Criteria for Surface Water Quality Classifications Florida Department of Environmental Prot ection (FDEP). (2008). 62302.530: Criteria for Surface Water Quality ClGeiger W. F. (1987). Flushing effects in combined sewer runoff. Proceedings of the 4th International Confer ence on Urban Storm Dr ainage, Lausanne, Switzerland. pp. 40-46. Geiger, W.F. Flushing effects in combined sewer systems. (1987). Proceedings of the 4th International Conference on Urban Storm Dr ainage, Lausanne, Switzerland. pp. 40-46.

PAGE 99

99 Gilbert, J.K., and Clausen, J.C. (2006). Stormwa ter runoff quality and quantity from asphalt, paver, and crushed stone driveways in Connecticut Water Research (40) pp. 826-832. Glenn III, D. W., Liu, D., and Sansalone, J.J. (2001). Influence of highway runoff chemistry, hydrology, and residence time on nonequilibrium pa rtitioning of metals. Implications for treatment at the highway shoulder. Transportation Research Record 1755. pp. 129-140. Glenn III, D. W. and Sansalone, J. J. (2002). A ccretion and partitioning of metals associated with snow exposed to urban traffic and winter storm maintenance activities II. Journal of Environmental Engineering 128(2), pp. 167-186. Gnecco, I., Berretta, C., Lanza, L.G., and La Ba rbera, P. (2005). Storm water pollution in the urban environment in Genoa, Italy. Atmospheric Research 77, pp. 60-73. Gupta, K. and Saul, A.J. (1996) Specific Relatio nships for the First Flush Load in Combined Sewer Flows. Water Resources 30 (5), pp 1244-1252. Hamilton, R. S., and Harrison, R. M. (1991). Highway Pollution Elsevier, New York. Hallenbeck, W.H. (1984). Human Health Effects of Exposure to Cadmium. Cellular and Molecular Life Sciences: 40 (2), pp. 136-142. Heaney, J. P., and Huber, W.C. (1984). Na tionwide assessment of urban runoff impact on receiving water quality, Water Resources Bulletin 20(1), 35-42. Herngren, L., Goonetilleke, A., and Ayoko, G.A. (2005). Understanding metal and suspended solids relationships in urban stor mwater using simulated rainfall. Journal of Environmental Management 76, pp. 149-158. Hogland, Wm. and Berndtsson, R. (1983). Quantita tive and qualitative characteristics of urban discharge to small river basins in the south west of Sweden. Nordic Hydrology, pp. 155166. Hope, D., Naegeli, M.W., Chan, A.H., and Grimm, N.B. (2004). Nutrients on asphalt parking surfaces in an urban environment. Water, Air, and Soil Pollution (4). pp. 371-390 Huber, W. (1993). Contaminant transport in surface water. D.R. Maidment, Ed., Handbook of Hydrology McGraw-Hill Inc., New York, N.Y., 14.1 14.50. Huber W., Nelson P., Eldin N., Williamson K., and Lundy J. (2001). Environmental impact of runoff from highway construc tion and repair m aterials. Transportation Research Record 1743, 1-9. Karouna-Reiner, N.K. and Sparling, D.W. (2001). Relationships between ambient geochemistry, watershed land-use and trace metal concentrations in aquatric invertebrates living in stormwater treatment ponds. Environmental Pollution 112. pp. 183-192.

PAGE 100

100 Korenromp, R.H.J., Hollander, J.C.Th., (1999). Di ffuse emissions of zinc due to atmospheric corrosion of zinc and zinc coated (galva nised) materials. TNO-MEP; Apeldoorn, Niederlande. Lee, J.H., Bang, K.W., Ketchum, L.H., Choe, J.S., a nd Yu, M.J. (2002). First flush analysis of urban storm runoff. The Science of the Total Environment (293). pp. 163-175. Li, L., Yin, C., He, Q., and Kong, L., (2007). First flush of storm runof f pollution for an urban catchment in China. Journal of Environment Sciences (19). pp. 295-299. McBride, M. (1994). Environmental chemistry of soils Oxford University Press, New York. pg. 126 Mersch, J. and Pihan, J.C. (1993). Simultane ous assessment of environmental impact on condition and trace metal availability in zebra mussels Dreissona polymorpha transplanted in Wiltz, River, Luxembourg. Archives of Environmental Contamination and Toxicology: 23 (3). pg. 353 Metcalf, L and Eddy, H. (1916) American Sewerage Practice Volu me III: Disposal of Sewage McGraw-Hill Inc. New York, NY 877 pp Morisawa, M., and LaFlure, E. (1979). H ydraulic geometry, stream equilibrium and urbanization. In Rhodes, D. D., and Williams, G. P. (Ed.) Adjustments of the fluvial systems. George Allen & Unwin (Publishers) Ltd. pp. 333-350. Mytyk, N. R. and J.J. Delfino. (2004). Evaluation of nitrate data in the Ocklawaha River Basin, Florida (1953 2002). Journal of the American Water Resources Association 40 (4), pp. 913-924. NAS (1972). Water Quality Criteria, A Report of the Committee on Water Quality Criteria. National Academy of Science, National A cademy of Engineering, Superintendent of Documents, Government Printing Office, Washington, D.C. Needleman, H.L. and Bellinger, D. (1991). The Health Effects of Low Level Exposure to Lead. Annual Review Public Health (12) pp. 111-140 Novotny, V., and Olem. (1994). Water Quality: pr evention, identification, and management of diffuse pollution. Van Nostrand Re inhold, New York, New York, USA. Pitt, R. Maestrae, A., and Morquecho R. (2004). The National Stormwater Quality Database (NSQD Version 1.1). Proceedings of American Society of Civil Engineers World Water and Environmental Resources 2004 Congress Salt Lake City, UT, pp. 35-40. Pizzarro, F., Olivares, M., Vauy, R ., Contreras, P., Rebelo, A., and Gidi, V. (1999). Acute Gastrointestinal Effects of Graded Le vels of Copper in Drinking Water. Environmental Health Perspectives : 107 (2). pp. 117-121.

PAGE 101

101 Rushton, B.T. (2000). Broadway Outfall Stormwater Retrofit Project, Phase II Monitoring CDS Unit and Constructed Pond Final Report in Progress. Ruston, B.T. (2002).Treatment of Stormwater Runoff From an Agricultural Basin by a WetDetention Pond in Ruskin, Florida Final Report Rushton, B. T. and Hastings, R. (2001). Florid a Aquarium Parking Lot, A Treatment Train Approach to Stormwater Management, Final Report. Sansalone, J.J. (2001). Physical and Chemi cal Nature of Urba n Storm Water Runoff Pollutants. Wet Weather Flow in the Urban Wate rshed. Technology and Management Chapter 3 Sansalone, J.J. and Buchberger, S.G. (1997). P artitioning and First Flus h of Metals in Urban Roadway Stormwater. Journal of Environmental Engineering pp. 135-143. Sansalone J. J., Koran, J. M., Smithson, J. A., and Buchberger, S. G. (1998). Physical characteristics of urban roadway solids transported during rain events. Journal of Environmental Engineering. 124(5), pp. 427-440. Sansalone, J.J., and Cristina, C.M. (2004) First Flush Concepts for Suspended and Dissolved Solids in Small Impervious Watersheds. Journal of Environmental Engineering ASCE, Vol. 130, No. 11 pp. 1301-1314. Sansalone J. and *Glenn D., M etal Distributions In Soils Receiving Urban Pavement Runoff And Snowmelt. Water Environment Research 79 (7): 736-752, July 2007. Sansalone J. J., and Ying, G. (2008). Granulomet ric relationships for ur ban source area runoff as a function of hydrologic event cl assification and sedimentation. Water, Air, and Soil Pollution 193 (1-4). Sheng, Y., Ying, G., and Sansalone, J. (2008). Di fferentiation of transpor t for particulate and dissolved water chemistry load indices in rainfall-runoff from urban source area watersheds. Journal of Hydrology (361). pp. 144-158. Soil Survey Staff, Natural Resources Conserva tion Service (NRCS), United States Department of Agriculture, (1997). Offici al series descriptions. http://soils.usda.gov/technical/classificatio n/osd/index.html. Accessed 15 June 2008. Lincoln, NE. Stahre, P., and Urbonas B. (1990). Stormwater Detention Prentice-Hall, Inc. Englewood Cliffs, N.J., pp. 1-338. Tchobanoglous, G., Burton, F. L., and Stensel, H. D. (2003). Wastewater Engineering: Treatment, Disposal, and Reuse 4th Ed., Metcalf and Eddy, Inc. McGraw Hill, Inc., New York.

PAGE 102

102 Thornton, R.C. and Saul, A.J. (1986), Some Quality Characteristics of Combined Sewer Flow. Public Health Engineering Vol. 24, pp 35-38. Tiefenthaler, L. L., Stein, E.D., and Schiff, K. C. (2008). Watershed and land use-based sources of trace metals in urban storm water. Environmental Toxicology and Chemistry 27(2), pp. 277-287. Tessier, A., Campbell, P.G.C., Auclair, J.C., a nd Bisson M. (1984). Trace Metals in Sediments and their accumulation in the tissues of the freshwater Mollusc Elliptio complanata in a Mining Area. Canadian Journal of Fisheries and Aquatic Sciences : 41 (10). pp. 14631472 Tong, S., Schimding, Y.E., and Prapamontol, T. (2000). Environmental Lead Exposure: a Public Health Problem of Global Dimensions. Bulletin of the World Health Organization: 78 (9) Turer D., Maynard B.J., and Sansalone J.J., Met al Contamination In So ils Of Urban Highways: Comparison Between Runoff And Soil Concentrations, J. of Water, Air and Soil Pollution, 132 (3), 214-230, December 2001. United States Department of Commerce. Na tional Oceanic & Atmospheric Administration. Annual Climatological Summ ary. http://cdo.ncdc.noaa.gov/ ancsum/ACS?stnid=20001598. Accessed 3 October 200 8. Asheville, NC. United States Environmental Protection Agency (USEPA). (2000). Establishment of numeric criteria for priority toxic pollutants for the stat e of California rule. Fe deral Register 65(97), 40CFR Part 131. pp. 31682-31718. United States Environmental Protection Agency (USEPA). (1994). Determination of trace elements in waters and wastes by inductivel y coupled plasma-mass spectrometry, revision 5.4. EPA method 200.8. pp. 1-5. United States Protection Agency (USEPA). ( 1999). National recommended water quality criteria Correction. EPA 822-Z-99-001. Office of Water, USEPA, Washington, D.C. Vorreiter and Hickey (1994). Incidence of the first flush ph enomenon in catchments of the Sydney region. National Conference Publicatio n Institute of Engineers (3). pp. 359-64. Walsh, C.T., Sandstead, H.H., Prasad, A.S., Newb erne, P.M., and Fraker, P.J. (1994). Zinc: Health Effects and Research Priorities for the 1990s. Environmental Health Perspectives : 102 (2). pp. 5-46 Wanielista, M.P. and Yousef, Y.A. (1993). Stormwater Management John Wiley and Sons, Inc., New York, N.Y. pp. 1-579. Weston Solutions, Inc. (2005). Los Angeles C ounty 1994-2005. Integrated receiving water impacts report.

PAGE 103

103 Ying, G. and Sansalone J. J. (2008). Granulometric relationships for urban source area runoff as a function of hydrologic event cl assification and sedimentation. Water, Air, and Soil Pollution 193 (1-4).

PAGE 104

104 BIOGRAPHICAL SKETCH Ki m Munksgaard was born to Diane and Don Munksgaard in 1984. She was raised in Coral Springs, Florida, and attended Coral Sp rings High School. After high school, she attended the University of Florida and pursued her bachel ors degree in environmental engineering. She graduated magna cum laude in December 2007. Through the BS/ME program offered at the University of Florida, she pursued the Master of Engineering degree in environmental engineering sciences, with a sp ecialization in stormwater. The degree was awarded in May 2009.