Proliferation of microalgae and enterococci in the Lake Okeechobee, St. Lucie, and Loxahatchee watersheds

Journal Pre-proof Proliferation of microalgae and enterococci in the Lake Okeechobee, St. Lucie, and Loxahatchee watersheds E. Kelly, M. Gidley, C. Sinigalliano, N. Kumar, L. Brand, R.J. Harris, H.M. SoloGabriele PII:

S0043-1354(19)31218-7

DOI:

https://doi.org/10.1016/j.watres.2019.115441

Reference:

WR 115441

To appear in:

Water Research

Received Date: 31 July 2019 Revised Date:

20 December 2019

Accepted Date: 22 December 2019

Please cite this article as: Kelly, E., Gidley, M., Sinigalliano, C., Kumar, N., Brand, L., Harris, R.J., SoloGabriele, H.M., Proliferation of microalgae and enterococci in the Lake Okeechobee, St. Lucie, and Loxahatchee watersheds, Water Research (2020), doi: https://doi.org/10.1016/j.watres.2019.115441. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.

Graphical Abstract

Concurrent algae blooms and enterococci exceedances

1

Proliferation of Microalgae and Enterococci in the Lake Okeechobee, St. Lucie, and

2

Loxahatchee Watersheds

3 4

E Kelly1,2,3, M Gidley3,5,6, C Sinigalliano3,5, N Kumar7, L Brand3,4, RJ Harris8, HM Solo-Gabriele1,2,3

5

1

University of Miami Leonard and Jayne Abess Center for Ecosystem Science and Policy, Coral Gables, FL, USA

6

2

University of Miami Department of Civil, Architectural and Environmental Engineering, Coral Gables, FL, USA

7

3

NSF NIEHS Oceans and Human Health Center, Rosenstiel School of Marine and Atmospheric Science, University of Miami,

8

Miami, FL, USA

9

4

University of Miami Department of Marine Biology and Ecology, Rosenstiel School of Marine and Atmospheric Science

10

(RSMAS), Miami, FL, USA,

11

5

12

Environmental Microbiology, Miami, USA

13

6

University of Miami Cooperative Institute for Marine and Atmospheric Studies (CIMAS), Miami, USA

14

7

University of Miami Department of Public Health Sciences, Division of Environment & Public Health, Miami, FL, USA

15

8

Loxahatchee River District, Jupiter, FL, USA

National Oceanic and Atmospheric Administration (NOAA) Atlantic Oceanographic and Meteorological Laboratory (AOML)

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

For consideration for potential publication in: Water Research

Version Dated: December 20, 2019

Corresponding author: Helena Solo-Gabriele, Ph.D. e-mail address: [email protected]

1

37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

Present/permanent address: E. Kelly: 1365 Memorial Drive, Ungar Building, 2nd Floor, Suite 230F, Coral Gables, FL 33146, USA M. Gidley: University of Miami Cooperative Institute for Marine and Atmospheric Studies, 4600 Rickenbacker Causeway, Miami, FL 33149 USA C. Sinigalliano: NOAA Atlantic Oceanographic and Meteorological Laboratory, Ocean Chemistry and Ecosystems Division, 4301 Rickenbacker Causeway, Miami, Florida 33149 USA N. Kumar: University of Miami Department of Public Health Sciences, Division of Environment & Public Health # 1063, Clinical Research Building 1120 NW 14th St., Miami, FL 33136 USA L. Brand: University of Miami Department of Marine Biology and Ecology, Rosenstiel School of Marine and Atmospheric Science (RSMAS), Miami, FL, USA R.J. Harris: Loxahatchee River District, 2500 Jupiter Park Drive, Jupiter, FL 33458 USA H.M. Solo-Gabriele: 1251 Memorial Drive, McArthur Engineering Building, Room 252, Coral Gables, FL 33146 USA

Abstract This study is an analysis of relationships between microalgae (measured as chlorophyll a)

63

and the fecal indicator bacteria, enterococci. Microalgae blooms and enterococci exceedances

64

have been occurring in Florida’s recreational waterways for years. More recently, this has

65

become a management concern as microalgae blooms have been attributed to potentially toxic

66

cyanobacteria, and enterococci exceedances link to human infection/illness. Since both the

67

microalgal blooms and bacterial exceedances occur in regions that receive managed freshwater

68

releases from Lake Okeechobee, we hypothesized that both the blooms and exceedances are

69

related to excess nutrients from the lake. Two experimental sites, on Lake Okeechobee and the

70

St. Lucie River (downstream of the lake), plus a control site on the Loxahatchee River (which

71

does not receive lake flow) were evaluated. The hypothesis was evaluated through three study

2

72

components: 1) analysis of available long-term data from local environmental databases, 2) a

73

year-long monthly sampling for chlorophyll a, enterococci, nutrients, and physical-chemical

74

data, and 3) microcosm experiments with altered water/sediment conditions. Results support the

75

hypothesis that excess nutrients play a role in both chlorophyll a and enterococci levels. For the

76

St. Lucie River, analyses indicate that chlorophyll a correlated significantly with total Kjeldahl

77

nitrogen (TKN) (R2=0.30, p=0.008) and the strongest model for enterococci included nitrate-

78

nitrite, TKN, total phosphorus (TP), orthophosphorus, and turbidity in our long-term analysis

79

(n=39, R2=0.83, p≤0.001). The microcosm results indicated that chlorophyll a and enterococci

80

only persisted for 36 hours in water from all sources, and that sediments from Lake Okeechobee

81

may have allowed for sustained levels of chlorophyll a and enterococci levels. Overall

82

similarities were observed in chlorophyll a and enterococci relationships with nutrient

83

concentrations regardless of a Lake Okeechobee connection, as underscored by a study of flow

84

out of the lake and downstream areas. This suggests that both nutrient-rich lake water and

85

untreated surface water runoff contribute to blooms and exceedances in southeast Florida.

86

Keywords: Enterococci, cyanobacteria, chlorophyll, blue-green algae, total Kjeldahl nitrogen,

87

phosphorus

88 89

1.0 Introduction

90

In south Florida, there has been an abundance of coastal cyanobacteria blooms and

91

exceedances of the fecal indicator bacteria (FIB) enterococci. In 2018 a confirmed cyanobacteria

92

bloom resulted in a state of emergency issued for seven counties across the state of Florida which

93

lasted from July until November 2018 (FDEP 2018). During the same period, enterococci

94

exceedances, instances when the number of enterococci bacteria was higher than 70 colony

3

95

forming units (CFU) per 100 mL occurred at beaches in these same counties (FDOH 2018.)

96

While some studies have included both chlorophyll a and FIB measurements, rarely have they

97

been evaluated simultaneously under the same monitoring program.

98

Cyanobacteria are a type of microalgae referred to as “blue-green algae”, due to their

99

color, but are technically bacteria that use photosynthesis (Parmar et al. 2011). There are various

100

groups and species of microalgae present across Florida but cyanobacteria, a majority of which

101

have historically been identified as Microcystis spp., typically dominate the microalgal blooms in

102

lakes and rivers throughout the region (Burns et al. 2008). These microalgae are a concern since

103

they have the ability to produce toxins, such as cyanotoxins, and the most commonly

104

encountered microcystin, which can be an irritant to humans and can result in hemorrhage of the

105

liver (see Dawson 1998). Cyanobacteria are monitored through the Florida Department of

106

Environmental Protection (FDEP) Algal Bloom Monitoring and Response program, along with

107

the Florida Fish and Wildlife Conservation Commission (FWC). These agencies identify the

108

species of microalgae, the concentrations and test for toxicity. The Florida Department of Health

109

(DOH) determines if an identified bloom presents a risk to human health and issues beach

110

advisories (FDEP 2018).

111

Bacterial exceedances, particularly FIBs such as enterococci, are monitored at coastal

112

beaches, some of which are downstream of rivers and lakes. The beaches are monitored for FIBs

113

by the DOH’s Florida Healthy Beaches Program (FHBP), a FIB monitoring database that

114

extends from 2000 to the current date. Water quality parameters including chlorophyll a and

115

nutrient levels, as well as flows of freshwater in lakes and rivers in Florida, such as Lake

116

Okeechobee, the St. Lucie River, and the Loxahatchee River are recorded through the efforts of

117

the South Florida Water Management District (SFWMD), the US Army Corps of Engineers

4

118

(USACE), and as part of the Comprehensive Everglades Restoration Program (CERP 2019).

119

Water quality in the Loxahatchee River, a designated Wild and Scenic River, is also monitored

120

by the Loxahatchee River District (LRD) (see Stoner and Arrington 2017).

121

Lake Okeechobee waters discharge to the St. Lucie River to the east of the lake through

122

the St. Lucie Canal to control flooding in the region near the lake. These discharges are cited by

123

researchers (Metcalf et al. 2018, Doering 1996) and citizens as the cause of the cyanobacterial

124

blooms and enterococci exceedances; others implicate septic tanks (Lapointe et al. 2015,

125

Lapointe 2012). Research by Donahue et al. 2017 further indicated that Florida beaches with

126

rivers within a 600 m distance of sampling sites had statistically higher FIB exceedances

127

compared to those without riverine influence. Two of the highest exceedances in the state were

128

sampled at Roosevelt Bridge (on the St. Lucie River, connected to Lake Okeechobee), and

129

Dubois Park (on the Loxahatchee River, not connected to Lake Okeechobee) (Donahue et al.

130

2017). Given the observations of cyanobacteria blooms in Lake Okeechobee (Havens et al. 2003)

131

and the St. Lucie River (Kramer et al. 2018), as well as the enterococci exceedances at coastal

132

beaches downstream, we hypothesized that water from Lake Okeechobee may be contributing to

133

cyanobacteria blooms and elevated enterococci within both Lake Okeechobee and the St. Lucie

134

River. Furthermore, we anticipated that the drivers of both the blooms and enterococci

135

exceedances may be related.

136

The objectives of this study were to evaluate the correlations between chlorophyll a

137

(measure of microalgae biomass) and enterococci with surface water quality measures, and to

138

conduct experiments to evaluate the influence of the water’s source on bacterial abundance and

139

algal biomass. To explore this, we (1) analyzed long-term data on selected water quality and

140

environmental measures, (2) conducted a year-long monthly field sampling, and (3) completed

5

141

microcosm studies in the laboratory. Of note, a confirmed cyanobacteria bloom, which resulted

142

in a state of emergency declaration for seven Florida counties, occurred in our study areas east of

143

Lake Okeechobee in Palm Beach and Martin Counties during the last three months of the year-

144

long field study (June-August 2018; See Supplemental Table 2). The year-long monthly field

145

sampling provides a unique data set prior to the 2018 cyanobacteria bloom (confirmed as

146

Microcystis aeruginosa).

147 148 149

2.0 Study Site Description Monthly field sampling sites for the year-long study included Canal Point, on Lake

150

Okeechobee; Roosevelt Bridge, east of the lake on the St. Lucie River; and Dubois Park, on the

151

Loxahatchee River, which is east of the lake and south of the St. Lucie River, and does not

152

receive discharges from Lake Okeechobee (Figure 1; GPS coordinates in Supplementary Table

153

S1). The Canal Point site was selected as it was the closest public access point to the lake,

154

located within 250 m of water control structure S352 and within 13 km of S308, where water

155

quality data were regularly collected by SFWMD (Figure 1). Roosevelt Bridge is located 110 m

156

east of a SFWMD water control structure (SE03) and 11 km east of water control structure S80

157

(Figure 1). Thus, data from sites S352, S308, and Canal Point were collected within Lake

158

Okeechobee; S80, SE03 and Roosevelt Bridge within the St. Lucie River; and Dubois Park

159

(control site, no connection to Lake Okeechobee) collected within the Loxahatchee River (Fig.

160

1). Water quality and hydrologic data collected by SFWMD were used in the long-term analysis,

161

as was enterococci and environmental data collected by DOH.

162 163

3.0 Methods

6

164

The study was conducted in three parts: (1) long-term (2000–2018) data trend analysis

165

for chlorophyll a, enterococci, and nutrients, along with an evaluation of physical-chemical data,

166

hydrologic measures, and environmental conditions; (2) a year-long field sampling campaign

167

from September 2017 until August 2018 documenting chlorophyll a, pheophytin, enterococci,

168

and nutrients, along with an evaluation of physical-chemical data, and environmental conditions;

169

(3) and a series of four laboratory-based microcosm studies of chlorophyll a and enterococci

170

using different sources of water and/or sediments.

171 172 173

3.1 Long-term data and analysis Long-term data were combined from two primary databases: SFWMD’s DBHYDRO

174

environmental database, and the DOH’s FHBP. DBHYDRO was used for chlorophyll a, nutrient,

175

physical-chemical and hydrologic data, whereas the DOH-FHBP was used for enterococci and

176

environmental data collected at the time of DOH sample collection. As the databases are part of

177

different agencies and are used for different purposes, not all of the variables, nutrients, physical-

178

chemical, or hydrologic data are available throughout the entire period of study. Most notably,

179

chlorophyll a could not be used for the multiple regression analysis due to the data gaps. No

180

DOH enterococci data was available for areas within Lake Okeechobee. Thus, Canal Point could

181

not be included in the long-term evaluation of DOH enterococci.

182 183

3.1.1 DBHYDRO nutrient and hydrologic data and analysis

184

DBHYDRO is an online database updated by the SFWMD and used to store hydrologic

185

and water quality data (www.sfwmd.gov). From DBHYDRO, data for chlorophyll a (and its

186

degradation product pheophytin), nutrients (nitrate-nitrite, total Kjeldahl nitrogen [TKN],

7

187

orthophosphorus, phosphorus [TP]), physical-chemical data (turbidity, water temperature, pH,

188

salinity, and dissolved oxygen [DO]) and hydrologic data (flow, gate, stage, rain, gage height)

189

were obtained. These data were collected from the monitoring stations within Lake Okeechobee

190

(S352 and S308), along the St. Lucie Canal (S80), and within the St. Lucie River (SE03). They

191

were evaluated using multiple linear regressions. The dependent variables, namely chlorophyll a,

192

pheophytin, and enterococci (DOH) were regressed on independent variables (specific nutrients,

193

physical-chemical measures, and hydrologic conditions) to identify which combination of

194

independent predictor variables significantly (p≤0.05) influenced the dependent variables.

195

Exploratory factor analysis (principal axis factoring analysis) was used to both identify

196

underlying covariance between predictor variables and select independent variables for model

197

inclusion (see Costello et al. 2005). Factor analysis was accomplished through PROC FACTOR,

198

where variables were scored (principal component method) and latent weighted (ordinary least

199

squares) co-variance extracted as a vector based on a Varimax–Orthogonal rotation (O’Rourke et

200

al. 2013). Results report the rate and direction of individual relationships (positive or negative

201

rate of change) as parameter estimates (i.e., regression coefficients) with associated p-values.

202

The final adjusted R2 values are used to account for the number of predictor variables included in

203

the models. All statistics were conducted using Statistical Analysis System (SAS) version 9.4.

204

Normality of data was assessed using residual plots and data were not transformed.

205 206

3.1.2 DOH enterococci and environmental data and analysis

207

The DOH collects and stores data through the FHBP and data is publicly available on the

208

FHBP website (Florida Department of Health 2016). The DOH FHBP data consisted of

209

enterococci and environmental data collected at the time of sampling by the DOH. These data

8

210

were separated into two types: quantitative (numeric) and qualitative (categorical). Quantitative

211

data included enterococci levels and quantitative environmental data (rainfall 24 hours before

212

collection, rainfall 3 days before collection, rainfall in the last week before collection, air

213

temperature, and water temperature). Qualitative data included categorized environmental data

214

(non-quantitative descriptions of weather conditions, tides, and current conditions). These data

215

were collected by DOH environmental specialists on a weekly basis within the St. Lucie River

216

(Roosevelt Bridge) and the Loxahatchee River (Dubois Park) and corresponded to conditions at

217

the time that enterococci samples were collected.

218

Statistical analysis for long-term DOH data included linear regression and analysis of

219

variance (ANOVA). Linear regression was used to evaluate enterococci with respect to

220

quantitative environmental data (rainfall, air temperature, water temperature), while ANOVA

221

using Statistical Analysis System (SAS)(SAS 9.4, SAS Institute, Cary, North Carolina) was used

222

to evaluate categorical environmental data (weather [sunny/cloudy/rainy], tidal conditions A

223

[high/medium/low], tidal conditions B [ebb/flood/slack], current direction A

224

[still/north/south/east/west], current direction B [along shore left to right/along shore right to

225

left/neither], and current strength [strong/moderate/weak]).

226 227

3.1.3 Analysis of DBHYDRO nutrient and hydrologic data and DOH enterococci data

228

Enterococci (DOH) were analyzed against nutrient and hydrologic data (DBHYDRO).

229

Statistical analysis was conducted using multiple regression and factor analysis as described

230

above. Additionally, within Lake Okeechobee (S308), some of the flows were recorded as

231

negative numbers, which corresponded to ‘back-pumping’, a practice in which water is pumped

232

out of the St. Lucie Canal into Lake Okeechobee, the opposite direction of normal flow. T-tests

9

233

were performed on the DOH Roosevelt Bridge enterococci and DBHYDRO S308 hydrologic

234

flow data (positive versus negative values) to identify significant differences (p ≤ 0.05) in

235

enterococci levels at Roosevelt Bridge based on flow direction. Building on this, hydrologic flow

236

data from S308 and S80 was also analyzed against Roosevelt Bridge enterococci and chlorophyll

237

a levels, along with other physical-chemical parameters from S80. Chlorophyll a, enterococci,

238

nutrients and turbidity levels were analyzed during periods where both S308 and S80 flow was

239

positive (measuring lake and downstream watershed contribution) and when flow from S308 was

240

zero and S80 was positive (measuring St. Lucie Canal watershed contribution only). These are

241

the only analyses in which DBHYDRO data and DOH data are evaluated against each other.

242 243 244

3.2 Year-long field study data and analysis To evaluate the potential drivers of the microalgae blooms and enterococci exceedances,

245

a year-long study of monthly field measurements was conducted at sites within Lake

246

Okeechobee (Canal Point), the St Lucie River (Roosevelt Bridge), and the Loxahatchee River

247

(Dubois Park).

248

During each monthly visit, surface (< 0.3 m) water samples were collected, and physical-

249

chemical measures and local meteorological conditions were recorded in the field. Water

250

samples were collected and then analyzed for chlorophyll a, pheophytin, enterococci, nutrients

251

(nitrate-nitrite, total Kjeldahl nitrogen [TKN], orthophosphorus, and phosphorus [TP]) and

252

turbidity. Brief descriptions of the analysis methods used for these measures are provided in the

253

supplemental text. Physical-chemical measurements of sampling depth, water temperature (ºC),

254

pH, salinity, and dissolved oxygen (DO) were recorded via field meter (Hydrolab Surveyor 4

255

with Hydrotech Compact Minisonde 4a). Meteorological conditions including air temperature

10

256

(ºC), wind, and humidity were recorded via the NavClock for iPhone App (Version 3.3.5, Split

257

Rail, Inc.), which reports these measurements from the nearest weather station based on GPS

258

location.

259

To assess relationships with rainfall, rainfall data were collected from the National

260

Oceanic and Atmospheric Administration (NOAA)’s National Centers for Environmental

261

Information (NCEI), formerly the National Climatic Data Center. NCEI rainfall data came from

262

weather stations near Canal Point (USC00081276), S308 (US1FLMT0016), Roosevelt Bridge

263

(USC00088620), and Dubois Park (USC00084461).

264

Analysis for the year-long monthly field sampling data included time-series analysis at

265

each site and multiple regression (following the same methods outlined above) of sampled

266

chlorophyll a and enterococci against individual nutrients and physical-chemical data. To mirror

267

the analysis of the rainfall data collected by DOH, and analyzed in this long-term study,

268

regression analysis was performed on the NCEI rainfall data using similar time periods (rainfall

269

amounts one month, one week, three days, and 24 hours before the collection dates) against the

270

chlorophyll a and enterococci data collected in the monthly field sampling study.

271 272 273

3.3 Microcosms The microcosm studies were designed to evaluate the impacts from Lake Okeechobee

274

water and sediment on microalgae (indicated by chlorophyll a) and enterococci. Surface water

275

samples were collected from Lake Okeechobee (Canal Point), the St. Lucie River (Roosevelt

276

Bridge) and the Loxahatchee River (Dubois Park). Sediments were collected from Lake

277

Okeechobee (Canal Point) and the Loxahatchee River (Dubois Park). Dubois Park water and

11

278

sediments were used as controls, as the Loxahatchee River is not connected to Lake Okeechobee

279

yet experiences high levels of enterococci.

280

Four laboratory-based microcosm experiments were conducted at the University of

281

Miami Department of Marine Biology and Ecology laboratory. All microcosms were conducted

282

without replication. For all microcosms, water was analyzed for chlorophyll a and enterococci

283

every 12 hours, for a 36-hour period following sample collection. Chlorophyll a and enterococci

284

were measured over time under two conditions: water only, and water with sediment. To isolate

285

the impacts from Lake Okeechobee, microcosms considered: 1) surface water only from Lake

286

Okeechobee (CP), the St. Lucie River (RB) and the Loxahatchee River (DP), collected June

287

2018, 2) surface water + deionized (DI) water (2:1 dilution) from Lake Okeechobee (Canal

288

Point) and the Loxahatchee River (Dubois Park), collected September 2017, 3) surface water +

289

sediment from Lake Okeechobee (Canal Point), collected November 2017, and 4) Lake

290

Okeechobee (Canal Point) surface water with either Lake Okeechobee sediments (Canal Point)

291

or Loxahatchee (Dubois Park) sediments, collected January 2018. Light and temperature were

292

designed for microalgae or enterococci growth. Conditions used for microalgae growth were 19

293

ºC with alternating light and dark cycles. For enterococci, growth conditions were 41ºC under

294

dark conditions (EPA 2009). Microcosms with sediment involved the lowering or raising of a

295

sediment tray into or out of the water every 12 hours to simulate wetting and drying cycles

296

(Supplementary Figure S1).

297

The first experiment, which evaluated water only at all three sites, consisted of six

298

individual 1 L clear plastic bottle microcosms. All six of the microcosms in the first experiment

299

were kept in algae room conditions (19 ºC, light/dark). Three microcosms consisted of water

300

bottles collected directly from Lake Okeechobee (Canal Point), the St. Lucie River (Roosevelt

12

301

Bridge) and the Loxahatchee River (Dubois Park). The other three microcosms consisted of

302

water from the same sites; the waters for these microcosms were filtered following the normal

303

procedure for total suspended solids (borosilicate filter, 1.5µm, nominal pore size).

304

The second experiment analyzed surface water from Lake Okeechobee (Canal Point) and

305

the Loxahatchee River (Dubois Park) considering pure samples and 2:1 DI water dilutions. Six

306

1L microcosms were created. One set (3 samples) of the collected water was maintained in the

307

algae culture room (19 ºC, light/dark); the second set of 3 was incubated under conditions

308

recommended for enterococci growth (41 ºC, dark). Microcosms in the algae room were housed

309

in clear glass bottles; those in the incubator conditions were housed in brown Nalgene sampling

310

bottles. All caps were vented. The water combinations included: (1) 500mL Dubois Park/500mL

311

Canal Point, (2) 500mL Dubois Park/500mL DI, and (3) 500mL Canal Point/500mL DI.

312

The third experiment evaluated only the Lake Okeechobee environment. The microcosms

313

were composed of water and sediment. Two microcosms were created, one for each condition

314

(algae culture room or enterococci incubator). A double tray system was developed to allow the

315

sediment to have periods of exposure as well as to be covered by water. Sediment was placed in

316

a tray (upper tray) at the bottom of a larger tray filled with water (lower tray). The upper tray was

317

raised or lowered at each sampling interval (every 12 hours), to simulate wetting and drying

318

cycles (Brakenhoff et al. 2018, Feng et al. 2015, Marchant et al. 2017) typically observed at the

319

coast due to tides. Measurements were taken before and after the tray was moved; measurements

320

taken after are designated by a .1 after the hour it was collected (e.g. 12.1).

321

The fourth experiment evaluated the effect of sediment source. Two microcosms were

322

created. One consisted of Lake Okeechobee water (Canal Point) and Loxahatchee River (Dubois

323

Park) sediments, and the other consisted of Lake Okeechobee (Canal Point) water and Lake

13

324

Okeechobee (Canal Point) sediments. Both microcosms were kept in algae culture room

325

conditions (19 ºC, light/dark) using the double tray system as in the third experiment with the

326

tray raised or lowered every 12 hours.

327

Lake Okeechobee sediment consisted of mixed rock/shell fill collected from the littoral

328

zone from Canal Point. Loxahatchee River sediment consisted of beach sand collected at Dubois

329

Park. Grain size distributions were measured for general categorization of sediment types using

330

the Standard Test Method for Particle-Size Analysis of Soils (ASTM 2007). See Supplementary

331

Table S4 for detailed results of grain size distributions.

332 333

4.0 Results

334 335 336 337

4.1 Long-term study data

338

against nutrient, physical-chemical, meteorological, and hydrologic data

339

4.1.1. DBHYDRO multiple regression of chlorophyll a, pheophytin, and enterococci

Analysis was conducted in two parts. The first part (Table 1) analyzed dependent

340

variables (chlorophyll a, pheophytin, and enterococci) against nutrients and physical-chemical

341

measures (i.e., independent predictor variables) at Lake Okeechobee (S308), the St. Lucie Canal

342

(S80) and the St. Lucie River (SE03) (Figure 1; Table 1). The second (Table 2) analyzed

343

enterococci levels at Roosevelt Bridge against the hydrologic data at Lake Okeechobee (S308)

344

and the St. Lucie Canal (S80). Among both data sets, the only significant correlations (p≤0.05)

345

were observed at the St. Lucie River (SE03) and the St. Lucie Canal (S80). None were observed

346

in Lake Okeechobee (S308).

347 348

At SE03, when evaluating chlorophyll a through multiple regression, the strongest model included pheophytin, TKN, water temperature and salinity (n=139, F=15.8, p≤0.0001, R2=0.30;

14

349

Table 1). Individually, chlorophyll a significantly correlated with pheophytin (parameter

350

estimate=1.12, p≤0.0001) and TKN (parameter estimate=6.92 p=0.008) as independent variables

351

(Table 1). The model for pheophytin at SE03 included chlorophyll a, nitrate-nitrite, TKN, total

352

phosphorus, water temperature, and salinity through multiple regression (n=139, F=12.0,

353

p≤0.0001, R2=0.32; Table 1). Individually, pheophytin was significantly correlated with

354

chlorophyll a only (parameter estimate=0.18, p≤0.0001; Table 1). The strongest enterococci

355

model at SE03 included chlorophyll a, pheophytin, nitrate-nitrite, TKN, total phosphorus,

356

orthophosphorus, turbidity, water temperature, salinity, and DO (n=39, F=19.2, p≤0.0001,

357

R2=0.83; Table 1). Individually, enterococci demonstrated significant correlations with nitrate-

358

nitrite, TKN, total phosphorus, orthophosphorus, and turbidity (individual p-values ≤0.05; Table

359

1).

360

For S80 for enterococci (DOH-Roosevelt Bridge data), the strongest multiple regression

361

model included TKN, total phosphorus, turbidity, pH, and DO (n= 49, F=13.8, p≤0.0001,

362

R2=0.57; Table 1). Individually, enterococci correlated significantly with TKN, and turbidity

363

(individual p-values ≤0.05; Table 1).

364

For the hydrologic data (Table 2) multiple regression models indicated that at S80

365

enterococci levels were correlated with flow, gate, stage, rain, and gage height (p=0.0005,

366

R2=0.23). Individually, enterococci correlated significantly with stage, rain, and gate height

367

(individual p-values ≤0.05). At S308 (Lake Okeechobee), the enterococci model included flow,

368

gate, stage, and rain, although the model was not significant (p=0.3, R2=0.008; Table 2) with no

369

individual variables correlating significantly with hydrologic data.

15

370

Overall, multiple regression results suggest that TKN is an important predictor of

371

chlorophyll a, pheophytin and enterococci. Enterococci also demonstrated significant

372

associations with total phosphorus, orthophosphorus, and turbidity.

373 374 375

4.1.2 DBHYDRO Lake Okeechobee (S308) and St. Lucie Canal (S80) Flow and DOH St. Lucie River (Roosevelt Bridge) Enterococci Data

376

Results from a T-test performed to distinguish differences in enterococci by flow

377

direction show that there was a statistically significant difference in enterococci levels (p≤0.001)

378

at Roosevelt Bridge when freshwater was flowing downstream (from Lake Okeechobee into the

379

St. Lucie River, mean enterococci 49.7 CFU/100mL, n=733) versus when it was reporting a

380

negative flow (from the St. Lucie River into Lake Okeechobee, mean enterococci 17.8

381

CFU/100mL, n=95). These findings suggest that flow from the lake may be a contributor to

382

enterococci at Roosevelt Bridge, serving as a direct bacterial source, indirectly promoting

383

conditions that allow enterococci to proliferate, or as a covariate of rainfall (see Figures 2a, 3a,

384

and 3b).

385

Building on this, hydrologic flow data from S308 and S80 was also analyzed against

386

Roosevelt Bridge chlorophyll a, enterococci, nutrient levels, and physical-chemical parameters.

387

The number of data points for chlorophyll a was small (a total of seven), so not all of the

388

analyses could be performed for this parameter. Hydrologic conditions were divided into four

389

categories depending upon whether flow in S308 or S80 was positive. These conditions

390

included: 1) S308 greater than zero and S80 flow at zero (no contribution to Roosevelt Bridge),

391

2) both S308 and S80 flow greater than zero (lake plus St. Lucie Canal watershed contribution),

392

3) both S308 and S80 flow at zero or less [only S308 has back pumping or negative flows] (no

393

contribution to Roosevelt Bridge), and 4) S308 flow is zero, S80 is greater than zero

16

394

(contribution from St. Lucie Canal watershed). Given the small sample size, chlorophyll a could

395

not be divided into all four categories; it was divided into categories 1 and 3. the means were

396

highest under condition 1 (7.50 µg/L), followed by condition 3 (5.80 µg/L). For enterococci, the

397

means were highest under condition 2 (74.3 CFU/100 ml), followed by condition 4 (44.7

398

CFU/100 ml), condition 1 (30.0 CFU/100 ml), and then condition 3 (26.0 CFU/100 ml). The

399

means for total phosphorus, nitrate-nitrite, TKN, and orthophosphorus followed the same pattern,

400

with their highest values under conditions 2 (both lake and watershed contributing) and 4

401

(watershed only contributing) (Table 3).

402

This was further analyzed through studies of condition 2 (measuring the combined Lake

403

Okeechobee and St. Lucie Canal watershed contribution) and condition 4 (measuring the St.

404

Lucie Canal watershed contribution alone). Chlorophyll a could not be analyzed, as there were

405

no data points for condition 2 or condition 4. For enterococci and orthophosphorus, condition 2

406

was higher; for TKN and total phosphorus, condition 4 was higher; for nitrate-nitrite, the means

407

were nearly the same in both conditions.

408

To further evaluate lake versus watershed contributions, condition 2 was evaluated

409

further in terms of the relative contributions from the lake versus the watershed. When flow

410

through S308 (Lake Okeechobee contribution) was higher than through S80 (St. Lucie Canal

411

watershed contribution) the enterococci mean was 58.9 CFU/100 ml. When flow through S80

412

was higher, the enterococci levels averaged 93.3 CFU/100 ml at Roosevelt Bridge. For nutrients

413

and physical-chemical parameters, the mean concentrations did not differ significantly in terms

414

of whether flows through S308 versus S80 were higher, with the exception of turbidity. The

415

turbidity when S308 had more flow was statistically higher than the turbidity (mean of 21.0

416

NTU) than when S80 had more flow (mean of 3.8 NTU).

17

417 418 419 420

4.1.3. Linear Regression and ANOVA for Analysis of DOH Enterococci and Environmental Data in the St. Lucie and Loxahatchee Rivers When quantitative environmental values were considered, linear regression showed that

421

enterococci at the St. Lucie River (Roosevelt Bridge) was significantly correlated with rainfall

422

during all periods tested, although the correlations were relatively weak with low R2 values

423

(R2≤0.04; Supplementary Table S5). Similarly, in the Loxahatchee River at Dubois Park,

424

correlations between enterococci and rainfall were significant, although R2 values were weak

425

(R2≤0.002). Enterococci at Roosevelt Bridge showed significant positive correlations with air

426

and water temperatures. Enterococci did not correlate significantly with temperatures at Dubois

427

Park (Supplementary Table S5).

428

When the categorical environmental variables were considered (Table 4), at Roosevelt

429

Bridge, enterococci levels did not show a significant difference when sorted by weather

430

(p≥0.05). At Dubois Park, enterococci levels, when sorted by different weather, were statistically

431

different (n=729, F-value 12.55, p≤0.0001), with a strong difference in enterococci levels

432

between weather observed to be sunny (1) versus cloudy (2) versus rainy (3) conditions (Table

433

4). Enterococci did not show significant differences with tidal conditions A or B at either site.

434

However, at Dubois Park, enterococci levels differed statistically by tidal conditions B (n=607,

435

F-value = 5.33, p-value = 0.005), with a strong difference between tide observed as ebb tide (1)

436

versus flood (2) and slack (3) tide. Enterococci levels did not significantly differ by current

437

direction A at either site. Current direction B was statistically significant at Roosevelt Bridge

438

(n=582, F-value= 5.7, p-value= 0.004) and at Dubois Park (F-value 5.3, p-value = 0.04). The

439

current strength categories showed no significant differences at either site.

18

440

Overall, the statistical variance with enterococci at the two sites demonstrated that

441

rainfall was associated with enterococci at both sites. Temperature and current direction B were

442

important at Roosevelt Bridge, while weather and tidal conditions B were factors significantly

443

influencing enterococci levels at Dubois Park (Table 4).

444 445 446 447

4.2 Year-Long Monthly Field Sampling Data and Analysis In the St. Lucie River, water quality parameters at the Roosevelt Bridge site changed

448

according to the tides. During flood tides the field meter readings of water quality parameters

449

were similar to those found in the Loxahatchee River at Dubois Park, whereas during ebb tides

450

the field meter readings were similar to parameters measured in Lake Okeechobee at Canal Point

451

(Supplementary Table S8). Meteorological data demonstrated that the wind and humidity

452

increased, and the air temperature dropped (Supplementary Table S6) toward the coast. For the

453

physical-chemical measures, salinity increased and turbidity decreased toward the coast. The

454

complete set of all of the observations is available in the supplement (Supplementary Tables S7

455

and S8).

456 457 458

4.2.1. Time-series analysis for each site In general, chlorophyll a, pheophytin, and enterococci decreased during the dry season

459

(November-April) and increased during the wet season (May-October), as shown in the

460

corresponding rainfall records of each site (Figures 2, 3, and 4). In Lake Okeechobee at Canal

461

Point (Figure 2), rainfall appeared to coincide with elevated chlorophyll a, pheophytin, and

462

enterococci from May through August; in March, TN, TKN, or phosphorus coincided with

463

increased enterococci. In the St. Lucie River at Roosevelt Bridge (Figure 3), chlorophyll a and

19

464

pheophytin demonstrated a peak during January 2018. Enterococci remained low (~100

465

CFU/100 ml or below) for most of the study, until June (1020 CFU/100 ml), when rainfall, TN,

466

TKN, TP, and orthophosphate all increased. Elevated levels were observed in all measures

467

shown from May until August. In the Loxahatchee River at Dubois Park (Figure 4), rainfall did

468

not appear to have as much of an impact on chlorophyll a, pheophytin, or enterococci at the

469

beginning of the study as it did at the end. Chlorophyll a was elevated for this site at the

470

beginning of the study), as was TN, TKN, and phosphorus. For the remainder of the field

471

sampling period, chlorophyll a and pheophytin remained at a low level (both around 3 µg/L or

472

below), while enterococci increased in March (200 CFU/100mL, up from 40 CFU/100mL in

473

February), along with increasing TN and TKN. Enterococci increased again in May (330

474

CFU/100mL, up from 30 CFU/100mL in April) coinciding with the wet season.

475

Overall, all concentrations tended to decrease from Lake Okeechobee (Canal Point) to the

476

St. Lucie (Roosevelt Bridge) to the Loxahatchee (Dubois Park) (Figures 2, 3, 4). Chlorophyll a,

477

pheophytin, and enterococci, generally increase with rainfall and nutrients.

478 479 480 481

4.2.2. Multiple regression of chlorophyll a, pheophytin, and enterococci against nutrient, physical-chemical, and meteorological data Multiple regression analysis of the data collected during the year-long field sampling

482

(Table 5) showed that for Lake Okeechobee at Canal Point, the model for chlorophyll a included

483

pH, salinity, and sample depth (n=12, F=3.9, p=0.05, R2=0.45). Independently, there were no

484

significant correlations between chlorophyll a and any of the independent variables (individual

485

p≥0.05). The model for pheophytin included chlorophyll a and enterococci, total phosphorus,

486

water temperature, salinity, DO, sample depth and wave height (n=12, F=9.8, p=0.04, R2=0.86).

20

487

Independently significant correlations were observed with chlorophyll a and DO (individual p≤

488

0.05). The model for enterococci included nitrate-nitrate, TKN, orthophosphorus, salinity, and

489

DO, but was not significant (n=12, F=1.2, p=0.4, R2=0.08). Enterococci did not correlate

490

significantly with any of the independent variables (individual p≥ 0.05).

491

In the St. Lucie River, at Roosevelt Bridge, the chlorophyll a model included pheophytin,

492

enterococci, TKN, pH, salinity and tide (n=11 F=63.9, p=0.0006, R2=0.99); independently, only

493

pheophytin was significant (parameter estimate=0.68 p=0.0003). The pheophytin model included

494

chlorophyll a, enterococci, nitrite-nitrate, orthophosphorus, and salinity (n= 8, F=32.6, p=0.03,

495

R2=0.96); independently, there were no significant associations. The multiple regression model

496

for enterococci included chlorophyll a, pheophytin, orthophosphorus, pH, salinity, and sample

497

depth (n=8, F=37.0, p=0.10, R2=0.99); independently, there were no significant associations.

498

In the Loxahatchee River, at Dubois Park, the model for chlorophyll a included total

499

phosphorus and salinity (n=12 , F=4.83, p=0.04, R2=0.41), the pheophytin model included

500

turbidity and salinity (n=12, F=2.28, p=0.16, R2=0.19), and the enterococci model included total

501

phosphorus, salinity, and sample depth (n=12, F=3.79, p=0.06, R2=0.43). Independently, there

502

were no significant associations for any of the variables (individual p≥0.05).

503

Overall, for the monthly data, salinity was a parameter identified for all models but as an

504

independent variable it was not significant. At the St Lucie River (Roosevelt Bridge), the model

505

for chlorophyll a included pheophytin and enterococci; the model for pheophytin included

506

chlorophyll a and enterococci; and the model for enterococci included chlorophyll a and

507

pheophytin.

508 509

4.2.3. Rainfall data and monthly field sampling for chlorophyll a and enterococci

21

510

For the analysis with NCEI rainfall data (Table 6), enterococci showed correlations with

511

all time scales (rainfall 24 hours before, 3 days before, week before, and month before).

512

Chlorophyll a and pheophytin showed significant correlations at the month time scales.

513

Correlations between chlorophyll a and pheophytin with rainfall at shorter time scales were not

514

consistent. Thus enterococci appears to respond to rainfall at multiple time scales both long and

515

short. Chlorophyll a and pheophytin respond a longer time scales.

516 517

4.3 Microcosms

518

Results from the microcosms suggest that the presence of sediment from Lake

519

Okeechobee (Canal Point) may have an influence on the proliferation of enterococci and may

520

contribute to increased chlorophyll a levels. In the first and second microcosm experiments

521

(Supplementary Figures S2 and S3), evaluating water from all of the sites, enterococci persisted

522

at about the same (normalized) level or dropped by 24 hours after the experiment started.

523

Chlorophyll a after 24 hours had increased. The high and persistent levels of chlorophyll was

524

expected, since this water was collected during a period of enterococci exceedance,

525

cyanobacteria bloom, and high turbidities in Lake Okeechobee and the St Lucie River. However,

526

Dubois Park, in the Loxahatchee River, did not have cyanobacteria.

527

In the third and fourth microcosm experiments (Figures 5 and 6), with Lake Okeechobee

528

sediment, enterococci persisted through 36 hours. When Dubois Park sediment was used,

529

enterococci concentrations did not persist and dropped after 12 hours (Figure 6). Chlorophyll a

530

and enterococci followed a cyclical pattern, in both the microcosm in the algae room and in the

531

incubator. This pattern may be in response to the raising and lowering of the sediment tray every

532

12 hours. The stirring up of the water and sediment may have caused the difference between the

22

533

lower measurement at 24 hours (made before the tray was moved) and the higher measurement

534

at 24.1 hours (made after the tray was moved).

535 536 537

5.0 Discussion This study was designed to analyze microalgae and enterococci in three watersheds in

538

association with environmental and water quality parameters that are known to be important to

539

both dependent variables. There have been few studies documenting the water

540

quality/environmental parameters of both enterococci and microalgae together. Most of what is

541

known about microalgae or about enterococci has been discovered through studies of either

542

microalgae or enterococci, along with the known drivers for that specific variable. Associations

543

between chlorophyll a, pheophytin and nutrients are well known (Dolman et al. 2012, Paerl et al.

544

2008, Havens et al. 2003, Kramer et al. 2018; Khare et al. 2019), and monitoring programs

545

typically analyze for these nutrients as part of their standard protocol. Additionally for

546

cyanobacteria, studies have also noted the influence of tides, currents, wind-driven waves and

547

shoreline conditions (Sootiens et al. 2018, Naja et al. 2017, Rosen and St. Amand 2015).

548

For enterococci, as consistent with other studies (Wright et al. 2011, Enns et al. 2012,

549

Barreras et al. 2019, Dila et al. 2018, McLellan and Roguet 2019, McLellan et al. 2015),

550

environmental factors such as tide, rainfall, and temperature have been associated with the levels

551

of enterococci as observed in the current study. Few studies exist that have evaluated

552

associations between environmental enterococci and nutrient levels (Cloutier et al. 2015). Most

553

studies that evaluate specifically the effects of nutrients on FIBs were conducted at wastewater

554

treatment plants, where nitrogen and phosphorus are also typically measured (LeChevallier et al

23

555

1991, Derry 2014). Here we found that rainfall, temperature, flows, and nutrients were found to

556

be significant independent predictor variables for both chlorophyll a and enterococci.

557

Most of what is known about growth factors for enterococci originates from the medical

558

field, with studies on environmental enterococci rapidly increasing. In evaluating enterococci

559

within the intestinal tract of mammals, Ramsey et al. 2014 noted that enterococci use the process

560

of phosphorylation, the addition of a phosphoryl group (PO3-) to provide adenosinetriphosphate

561

(ATP) to the cell. In a comparative genomic analysis of Enterococcus, some of the type strains

562

were found to have strain-specific genes for nitrogen metabolism and 18 core genes (and one

563

strain-specific gene) for phosphorus metabolism (Zhong et al 2017). This may indicate that

564

enterococci can use nutrients in the extraenteric environment, which seems to be in opposition to

565

findings that indicated they require complex nutrients (Lebreton et al. 2014). Enterococci may

566

also be protected in oligotrophic environments due to their thick peptidoglycan wall, and their

567

use of biofilm, both of which are known to offer protective barriers in extraenteric environments

568

(Chang et al. 2018).

569

The closest study of correlation between microalgae and enterococci simultaneously

570

revealed that enterococci “may be highly concentrated in plankton (which included

571

cyanobacteria) and associated particles, especially summer and fall months” (Mote et al 2012).

572

Other studies echo the association between enterococci and plankton, asserting that enterococci

573

survive because they adhere themselves to the plankton, (Maugeri et al 2004) which serve as

574

reservoirs for enterococci (Signoretto et al. 2004). Byappanahalli et al. (2012) recommended

575

further study of ecological factors on the survival of enterococci recognizing the influence of

576

organisms in controlling enterococci populations. Despite the similarities of potential drivers,

24

577

there have been few studies documenting the water quality/environmental parameters of both

578

FIB and microalgae together.

579

It is important to note that this study began immediately following Hurricane Irma, which

580

made landfall in south Florida in September 2017, producing large amounts of rainfall at the lake

581

and both rivers. Hurricane Irma may have impacted the watersheds examined in the current

582

study, and subsequently may have been an important factor in the cyanobacterial bloom and

583

enterococci exceedances measured during our year-long study in 2018. Increases in enterococci

584

levels immediately after hurricanes has been documented (Roca et al. 2019) but such increases

585

are not necessarily noticeable for cyanobacteria. For example after Hurricanes Katrina and Rita,

586

elevated levels of enterococci were noted in the Lake Pontchartrain area for a period of up to two

587

months after the storm whereas no evidence of a cyanobacterial bloom was noted (Sinigalliano et

588

al. 2007). In the longer term, the timing of the sampling study was very fortunate when it comes

589

to the evaluation of enterococci and chlorophyll a, as it coincided with a FDEP-confirmed 2018

590

cyanobacterial bloom and the declared state of emergency in southeastern Florida. In Spring

591

2018, during this study, the wet season began with heavy rains. The wet season, a time of

592

increased rain and high temperatures, has also proven to be a time of increased nutrients (e.g.,

593

Lapointe et al. 2012). Enterococci spikes were reported at both Roosevelt Bridge and Dubois

594

Park, following the rain, and blue-green algae (cyanobacteria) was documented by FDEP near

595

Roosevelt Bridge. This increase in dependent variables and the nutrients together during the wet

596

season was detected in the time-series analysis influencing both enterococci and cyanobacteria

597

levels.

598 599

Results from the microcosms suggested that the presence, and/or re-suspension of sediments/fine material from Lake Okeechobee may have contributed to the proliferation of

25

600

enterococci and Lake Okeechobee water to chlorophyll biomass. Chlorophyll is known to

601

maintain a diurnal cycle in response to light and darkness (Leypunskiy et al. 2017), yet we

602

observed the cycling of both chlorophyll and enterococci, in both the algae culture room and the

603

incubator (peaks at 0 and 24 hours; Fig. 5). It is difficult to determine whether the cycling is due

604

to circadian patterns, or the action of raising and lowering the sediment tray; which also occurred

605

every 12 hours and may have initiated resuspension. The importance of re-suspension has been

606

noted at recreational beaches, where enterococci proliferate in sediments and, with wave action,

607

are released from these sediments into the water column (Ferguson et al. 2005, Desmarais et al.

608

2002, Verhougstraete and Rose 2014, Whitman et al. 2014, Feng et al. 2016). Moreover, wind-

609

driven re-suspension has been previously noted in Lake Okeechobee (Maceina and Soballe 1990)

610

and can be a precursor to cyanobacteria blooms (Paerl 1988). Here, the Lake Okeechobee

611

sediments were classified as poorly graded gravels (see Supplemental Table S4), yet we

612

observed a fine black material present in the Lake Okeechobee sediments (field observations,

613

this study); in contrast to Dubois Park’s sandy sediments (Supplemental Table S4). Typically,

614

gravels can promote a buildup of fine organic material in pore spaces, increasing localized re-

615

suspension (Koiter et al 2015), whereas a low fraction of fine particles present in in sands can

616

stabilize sediments (Bartzke et al. 2013). A small fraction of fine material could have made an

617

impact in Lake Okeechobee water quality.

618 619 620

6.0 Conclusions The results from this study demonstrate that excess nutrients are associated with elevated

621

chlorophyll a and enterococci levels. While the correlations between nutrients, chlorophyll a,

622

pheophytin and enterococci were similar across sites, the driving physical-chemical,

26

623

environmental, and meteorological variables were site-specific. We found that enterococci levels

624

may be influenced by flow from Lake Okeechobee into the St. Lucie River, as seen through the

625

analysis of long-term hydrologic data. When flow at S308 was reversed, enterococci levels

626

dropped significantly at Roosevelt Bridge. At Dubois Park, we also observed the importance of

627

tidal conditions and short-term rainfall. This may indicate that chlorophyll a and enterococci

628

levels change based on a combination of the available nutrients and the physical-chemical,

629

environmental, and meteorological conditions at each site.

630

When hydrologic flow data from S308 and S80 was analyzed against Roosevelt Bridge

631

enterococci, and chlorophyll a, nutrient levels, and turbidity at S80, we found that the means for

632

almost every parameter measured was highest for condition 2, where both Lake Okeechobee and

633

the St. Lucie Canal had flows greater than zero (both lake and watershed contributions), and then

634

for condition 4, where only the St. Lucie Canal had flows greater than zero (watershed

635

contribution). This supports the hypothesis that the lake and downstream watershed contribute to

636

the chlorophyll a, enterococci, and physical-chemical parameter levels at receiving waters

637

downstream. The highly significant turbidity response when flows were high from the Lake

638

again implicates that both the lake and downstream watershed are contributing, perhaps

639

enhanced through fine sediment transport out of the lake and into the St. Lucie Canal.

640

Overall, similarities were observed in the relationships of chlorophyll a and enterococci

641

with nutrients, regardless of a Lake Okeechobee connection. This suggests that both nutrient-rich

642

lake water and untreated runoff contribute to the blooms and exceedances in southeast Florida.

643

These similarities in response to nutrients and hydrologic conditions may explain the coincidence

644

of both cyanobacteria blooms and bacterial exceedances observed during the summer of 2018

645

within the downstream areas of the St. Lucie River.

27

646

Future work is necessary to evaluate the possible connections between cyanobacteria and

647

enterococci, and to understand the biogeochemistry of areas that have blooms and exceedances.

648

These studies should focus on the evaluation and characterization of the nutrients specific to each

649

watershed, as well as the native sediment, organic matter, and local hydrodynamics. This should

650

include work on the effects of environmental conditions such as rainfall, and extreme weather

651

conditions such as hurricanes. Further studies on enterococci levels and the direction of flows

652

near the Roosevelt Bridge sampling area should also be included. These studies will provide

653

greater understanding as to whether sediments could be contributing to blooms and exceedances

654

in the St. Lucie River, especially after periods of heavy rainfall. An understanding of the

655

influence of weather events on blooms and exceedances may lead to the eventual forecasting of

656

blooms and/or exceedances.

657 658

Acknowledgments

659

This study was funded through the support of the Everglades Foundation, the University of

660

Miami’s Leonard and Jayne Abess Center for Ecosystem Science and Policy, and the University

661

of Miami Oceans and Human Health Center. Collaboration on this project was facilitated by the

662

University of Miami ULINK program. We wish to thank the South Florida Water Management

663

District (SFWMD) for their assistance with DBHYDRO and in identifying appropriate

664

monitoring stations. Special thanks also go to Loxahatchee River District for providing support

665

and field sampling supplies for the year-long study, along with supplies, training, equipment, and

666

analysis of nutrients.

667

28

668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711

References ASTM International, 2007. ASTM D422-63(2007)e2, Standard Test Method for Particle-Size Analysis of Soils (Withdrawn 2016). www.astm.org Barreras Jr H, Solo-Gabriele HM, Kelly EA, Kumar N. 2019. Assessment of local and regional strategies to control bacteria levels at beaches with consideration of impacts from climate change. Marine Pollution Bulletin, 138: 249-259. Bartzke G, Bryan KR, Pilditch CA, Huhn K, 2013. On the stabilizing influence of silt on sand beds. Journal of Sedimentary Research, 83(8): 691-703 Brakenhoff LB, Smit Y, Donker JJA, Ruessink, G, 2018. Tide‐induced variability in beach surface moisture: Observations and modelling. Earth Surface Processes and Landforms, 44: 317– 330 Burns J, 2008. Toxic cyanobacteria in Florida waters. In: Hudnell KH (Ed.), Cyanobacterial Harmful Algal Blooms: State of the Science and Research Needs. Advances in Experimental Medicine and Biology, vol. 619, vol. XXIV. pp. 127–137 Byappanahalli M, Nevers M, Korajkic A, Staley Z, Harwood V 2012. Enterococci in the Environment. Microbiology and Molecular Biology Reviews, 76: 685-706 Chang J, Wallace AG, Foster EE, Kim SJ 2018. Peptidoglycan composition analysis of Enterococcus faecalis biofilm by stable isotope labeling by amino acids in bacterial culture. Biochemistry, 57: 1274-1283 Cloutier DD, Alm EW, McLellan SL, 2015. Influence of land use, nutrients, and geography on microbial communities and fecal indicator abundance at Lake Michigan beaches. Applied and Environmental Microbiology, 81: 4904-4913 Comprehensive Everglades Restoration Project (CERP), 1999. https://www.saj.usace.army.mil/Missions/Environmental/Ecosystem-Restoration/CentralEverglades-Planning-Project/ Viewed 2/12/18 Costello AB, Osborne J, 2005. Best practices in exploratory factor analysis: four recommendations for getting the most from your analysis. Practical Assessment Research & Evaluation, 10: 1-9 Dawson RM, 1998. The toxicology of microcystins, Toxicon, 36: 953-62 Derry C, Attwater R, 2014. Regrowth of enterococci indicator in an open recycled-water impoundment. Science of the Total Environment, 486: 63-67

29

712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757

Desmarais TR, Solo-Gabriele HM, Palmer CJ, 2002. Influence of soil on fecal indicator organisms in a tidally influenced subtropical environment. Applied and Environmental Microbiology, 68(5): 1165-1172 Dila DK, Corsi SR, Lenaker PL, Baldwin AK, Bootsma MJ, McLellan SL, 2018. Patterns of host-associated fecal indicators driven by hydrology, precipitation, and land use attributes in Great Lakes watersheds. Environmental Science & Technology, 52(20): 1150011509 Donahue A, Feng Z, Kelly E, Reniers A, Solo-Gabriele H, 2017. Significance of beach geomorphology on fecal indicator bacteria levels. Marine Pollution Bulletin, 121(1-2): 160-167 Doering PH ,1996. Temporal variability of water quality in the St. Lucie Estuary, South Florida. Journal of the American Water Resources Association, 32: 1293–1306 Dolman AM, Rucker J, Pick FR, Fastner J, Rohrlack T, Mischke U, Wiedner C, 2012. Cyanobacteria and cyanotoxins: the influence of nitrogen versus phosphorus. PLoS ONE, 7(6): e38757 Enns AA, Vogel LJ, Abdelzaher AM, Solo-Gabriele HM, Plano LRW, Gidley ML, Phillips MC, Klaus JS, Piggot AM, Feng Z, Reniers AJHM, Haus BK, Elmir SM, Zhang Y, Jimenez NH, Abdel-Mottaleb N, Schoor ME, Brown A, Khan SQ, Dameron AS, Salazar NC, Fleming LE, 2012. Spatial and Temporal Variation in Indicator Microbe Sampling is Influential in Beach Management Decisions, Water Research, 46: 2237-2246 EPA Method 1600: Enterococci in Water by Membrane Filtration Using membraneEnterococcus Indoxyl-β-D-Glucoside Agar (mEI) December 2009 https://nepis.epa.gov Viewed 1/22/18 Feng Z, Reniers A, Haus BK, Solo-Gabriele HM, Wang JD, Fleming LE, 2015. A predictive model for microbial counts on beaches where intertidal sand is the primary source. Marine Pollution Bulletin, 94: 37-47 Feng Z, Reniers A, Haus BK, Solo-Gabriele HM, Kelly EK, 2016. Wave energy level and geographic setting correlate with Florida beach water quality. Marine Pollution Bulletin, 104(12): 54-60 Ferguson D, Moore D, Getrich M, and Zhowandai, M, 2005. Enumeration and speciation of enterococci found in marine and intertidal sediments and coastal water in southern California. Journal of Applied Microbiology, 99: 598-608. Florida Department of Environmental Protection (FDEP), 2018. Algal Bloom Sampling Results 2018. https://floridadep.gov/dear/algal-bloom/content/algal-bloom-sampling-results Viewed 3/16/19 Florida Department of Health, 2016. Florida Healthy Beaches Program. http://www.

30

758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803

floridahealth.gov/environmental-health/beach-water-quality/index.html. Viewed 1/22/18 Havens KE, Walker Jr. WW, 2002. Development of a total phosphorus concentration goal in the TMDL process for Lake Okeechobee, Florida (USA). Lake and Reservoir Management, 18(3): 227-238 Havens KE, James RT, East TL, Smith VH, 2003. N:P Ratios, light limitation, and cyanobacterial dominance in a subtropical lake impacted by non-point source nutrient pollution. Environmental Pollution, 122(3): 379-390 Khare Y, Naja GM, Stainback, GA, Martinez CJ, Paudel R, Van Lent T, 2019. A phased assessment of restoration alternatives to achieve phosphorus water quality targets for Lake Okeechobee, Florida, USA. Water, 11: 327 Koiter AJ, Owens PN, Petticrew EL, Lobb DA, 2015. The role of gravel channel beds on the particle size and organic matter selectivity of transported fine-grained sediment: implications for sediment fingerprinting and biogeochemical flux research. Journal of Soils and Sediments, 15(10): 2174–2188 Korajkic A, Badgley BD, Brownell MJ, Harwood VJ, 2009. Application of microbial source tracking methods in a Gulf of Mexico field setting. Journal of Applied Microbiology, 107: 15181527 Kramer BJ, Davis TW, Meyer KA, Rosen BH, Goleski JA, Dick GJ, Oh G, Gobler CJ, 2018. Nitrogen limitation, toxin synthesis potential, and toxicity of cyanobacterial populations in Lake Okeechobee and the St. Lucie River Estuary, Florida, during the 2016 state of emergency event. PLoS ONE, 13(5): e0196278. Lapointe BE, Herren L, Bedford B, 2012. Effects of hurricanes, land use, and water management on nutrient and microbial pollution: St. Lucie Estuary, Southeast Florida. Journal of Coastal Research, 28(6): 1345–1361 Lapointe BE, Herren L, Debortoli D, Vogel M, 2015. Evidence of sewage-driven eutrophication and harmful algal blooms in Florida’s Indian River Lagoon. Harmful Algae, 43: 82–102 Lebreton F, Willems RJL, Gilmore MS 2014. Enterococcus Diversity, Origins in Nature, and Gut Colonization. From Enterococci: From Commensals to Leading Causes of Drug Resistant Infection [Internet]. NCBI https://www.ncbi.nlm.nih.gov/books/NBK190427/ LeChevallier M, Schulz W, and Lee R, 1991. Bacterial Nutrients in Drinking Water. Applied and Environmental Microbiology, 57:857-862 Leypunskiy E, Lin J, Yoo H, Lee U, Dinner AR, Rust MJ, 2017. The cyanobacterial circadian clock follows midday in vivo and in vitro. eLife, 6, art. no. e23539 https://elifesciences.org/articles/26575

31

804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834

Maceina MJ, Soballe DM, 1990. Wind-related limnological variation in Lake Okeechobee, Florida. Lake and Reservoir Management, 6(1): 93-100

O’Rourke N and Hatcher L, 2013. A Step-by-Step Approach to Using SAS for Factor Analysis and Structural Equation Modeling (2nd Ed.). Cary, NC: SAS Press.

835 836

Paerl HW, 1988. Nuisance phytoplankton blooms in coastal, estuarine, and inland waters. Limnology and Oceanography, 33(2): 823-847

837 838 839

Paerl HW, Joyner JJ, Joyner AR, Arthur K, Paul V, O’Neil JM, Heil CA, 2008. Co-occurrence of dinoflagellate and cyanobacterial harmful algal blooms in southwest Florida coastal waters: dual nutrient (N and P) input controls. Marine Ecology Progress Series, 371: 143–153

840 841 842

Parmar A, Singh NK, Pandey A, Gnansounou E, Madamwar D, 2011. Cyanobacteria and microalgae: A positive prospect for biofuels. Bioresource Technology, 102(22): 10163-10172

Marchant HK, Ahmerkamp S, Lavik G, Tegetmeyer HE, Graf J, Klatt JM, Holtappels M, Walpersdorf E, Kuypers MMM, 2017. Denitrifying community in coastal sediments performs aerobic and anaerobic respiration simultaneously. The ISME Journal, 11: 1799–1812 Maugeri TL, Carbone M, Fera MT, Irrera GP, Gugliandolo C, 2004 Distribution of potentially pathogenic bacteria as free living and plankton associated in a marine coastal zone. Journal of Applied Microbiology, 97: 354-361 McLellan SL, Fisher JC, Newton RJ, 2015. The microbiome of urban waters. International Microbiology, 18: 141-149 McLellan SL, Roguet A, 2019. The unexpected habitat in sewer pipes for the propagation of microbial communities and their imprint on urban waters. Current Opinion in Biotechnology, 57: 34-41 Metcalf JS, Banack SA, Powell JT, Tymm FJM, Murch SJ, Brand LE, Cox PA, 2018. Public health responses to toxic cyanobacterial blooms: perspectives from the 2016 Florida event. Water Policy, 20(5): 919–932 Mote B, Turner J, Lipp E 2012. Persistence and Growth of the Fecal Indicator Bacteria Enterococci in Detritus and Natural Estuarine Plankton Communities. Applied and Environmental Microbiology, 78 (4): 2569-2577 Naja GM, Childers DL, Gaiser EE, 2017. Water quality implications of hydrologic restoration alternatives in the Florida Everglades, United States. Restoration Ecology, 25(S1): S48-S58

32

843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888

Ramsey M, Hartke A, Huycke M, 2014. The Physiology and Metabolism of Enterococci. From Enterococci: From Commensals to Leading Causes of Drug Resistant Infection [Internet]. NCBI https://www.ncbi.nlm.nih.gov/books/NBK190427/ Roca MA, Brown R, Solo-Gabriele HM, 2019. Fecal indicator bacteria levels at beaches in the Florida Keys after Hurricane Irma. Marine Pollution Bulletin, 138: 266-273. Rosen BH, St. Amand A, 2015. Field and laboratory guide to freshwater cyanobacteria harmful algal blooms for Native American and Alaska Native Communities: U.S. Geological Survey Open-File Report 2015–1164 Signoretto C, Burlacchini G, del Mar Lleo M, Pruzzo C, Zampini M, Pane L, Franzini G, Canepari P, 2004. Adhesion of Enterococcus faecalis in the Nonculturable State to Plankton is the Main Mechanism Responsible for Persistence of this Bacterium in both Lake and Seawater. Applied and Environmental Microbiology, 70: 6892-6896 Sinigalliano CD, Gidley ML, Shibata T, Whitman D, Dixon TH, Laws E, Hou A, Bachoon D, Brand L, Amaral-Zettler L, Gast R, Steward GF, Nigro OD, Fujioka R, Betancourt WQ, Vithanage G, Mathews J, Fleming LE, Solo-Gabriele HM, 2007. Microbial Landscape Measurements in Water and Sediments of the New Orleans Area in the Aftermath of Hurricanes Katrina and Rita. Proceedings of the National Academy of Sciences of the United States of America, 104(21): 9029-9034 Soontiens N, Binding C, Fortin V, Mackay M, Yerubandi RR, 2018. Algal bloom transport in Lake Erie using remote sensing and hydrodynamic modelling: Sensitivity to buoyancy velocity and initial vertical distribution. Journal of Great Lakes Research, 45(3): 556-572 Stoner EW, Arrington DA. 2017. Nutrient inputs from an urbanized landscape may drive water quality degradation. Sustainability of Water Quality and Ecology, 9: 136-150 Verhougstraete M, Rose JB, 2014. Microbial investigations of water, sediment, and algal mats in the mixed use watershed of Saginaw Bay, Michigan. Journal of Great Lakes Research, 40: 75-82 Whitman RL, Harwood VJ, Edge TA, Nevers MB, Byappanahalli M, Vijayavel K , Brandao J, Sadowsky MJ, Wheeler Alm E, Crowe A, Ferguson D, Ge Z, Halliday E, Kinzelman J, Kleinheinz G, Przybyla-Kelly K, Staley C, Staley Z, Solo-Gabriele HM, 2014. Microbes in beach sands: integrating environment, ecology and public health. Reviews in Environmental Science and Biotechnology, 13: 329–368 Wright ME, Solo-Gabriele HM, Abdelzaher AM, Elmir S, Fleming LE, 2011. The inter-tidal zone is the geographic location of elevated concentrations of enterococci. Water Science & Technology, 63.3: 542-549 Zhong Z, Zhang W, Song Y, Liu W, Xu H, Xi X, Menghe B, Zhang H, Sun Z, 2017. Comparative genomic analysis of the genus Enterococcus. Microbiological Research 196: 95105

33

Dependent Variables

Enterococci (DOH Roosevelt Bridge) n= 230, F=1.87, p=0.08, R2= 0.03

Enterococci (DOH Roosevelt Bridge) n=49, F=13.8, p≤0.0001, R2= 0.57

Enterococci (DOH Roosevelt Bridge) n=39, F=19.2, p≤0.0001, R2= 0.83

Chlorophyll a (DBHYDRO Roosevelt Bridge) n=139, F=15.8 p≤0.0001, R2= 0.30

Pheophytin (DBHYDRO Roosevelt Bridge) n=139, F=12.0, p≤0.0001, R2= 0.32

Independent Variables Parameter Estimate, p-value Measurement Location for S308, S80, Independent Variables Lake St. Lucie Canal Okeechobee Not included Not included Chlorophyll a (µg/L) Not included Not included Pheophytin (µg/L)

SE03, St. Lucie River

SE03, St. Lucie River

SE03, St. Lucie River

0.75, 0.70 -0.90, 0.73

Not included 1.12, ≤0.0001

0.18, ≤0.0001 Not included

Enterococci (CFU/100 mL) Not included 61.9, 0.06 Nitrate-nitrite (mg/L)

Not included Not included

Not included 304, 0.001

Not included Not included

Not included 1.71, 0.60

TKN (mg/L) Total Phosphorus (mg/L) Orthophosphorus (mg/L) Turbidity (NTU) Water Temperature (°°C)

-33.7, 0.10 -6.26, 0.95 Not included 0.27, 0.30 3.67, 0.06

57.3, 0.02 -170, 0.07 Not included 1.87, 0.0002 Not included

-73.8, 0.04 1345, ≤0.0001 -1320, ≤0.0001 -3.66, 0.001 1.63, 0.5

6.92, 0.008 Not included Not included Not included 0.11, 0.50

-0.64, 0.60 6.66, 0.06 Not included Not included 0.12, 0.10

pH Salinity (ppt) DO (mg/L)

11.6, 0.6 Not included 0.64, 0.90

-7.18, 0.647 Not included 0.21, 0.95

Not included -0.078, 0.9 4.97, 0.60

Not included 0.12, 0.20 Not included

Not included -0.010, 0.80 Not included

Table 1. Individual nutrient and physical-chemical measures included in multiple regression models for chlorophyll a, pheophytin, and enterococci. Not included=insufficient data to conduct multiple regression or comparison between dependent variable and itself omitted.

1

Dependent Variables

Independent Variables Measurement Location for Independent Variables Flow (cfs) Gate (feet) Stage (feet NGVD29) Rain (inches) Gage height (feet)

Enterococci DOH Roosevelt Bridge

Enterococci DOH Roosevelt Bridge n=98, F=1.19, n=74, F=5.05, p=0.3, R2=0.008 p=0.0005, R2=0.23 Parameter Estimate, p-value S308, S80, Lake Okeechobee St. Lucie Canal -0.0077, 0.4 -0.02, 0.4 8.12, 0.1 59.1, 0.5 4.84, 0.2 -308, 0.04 -0.40, 0.1 84.3, <0.0001 Not included 307, 0.04

Table 2. Hydrologic measures significantly correlating with enterococci using multiple regression by site. Insufficient chlorophyll a and pheophytin data at S308 and S80 to conduct multiple regression. Gate=height at which the gate is lifted to release water. Stage=height or elevation of water above an established datum plane (here NGVD29). Gage height=height of the water in the stream above a reference point.

2

Conditions

Chlorophyll a (µg/L)

Enterococci (CFU per 100 mL)

Total Phosphorus (mg/L)

Nitrate_Nitrite (mg/L)

1

7.5

29.96

0.11

0.06

2

N/A

74.26

0.16

3

5.8

26.01

4

N/A

44.73

Total Kjeldahl Nitrogen (mg/L)

Orthophosphorus (mg/L)

Turbidity (NTU)

0.92

0.05

7.43

0.2

1.17

0.09

15.32

0.11

0.1

0.72

0.04

5.97

0.27

0.2

1.37

0.07

20.08

Table 3. Means for Chlorophyll a, enterococci, and selected physical-chemical parameters.

3

Variable

Weather (sunny (1), cloudy (2), rainy (3))

Tidal Conditions A (high, medium, low)

Tidal Conditions B (ebb, flood, slack)

ANOVA, Categorical Variables Site Mean (CFU/100mL) Roosevelt Bridge 1=43.4 2=42.9 3=69.0 Dubois Park 1=31.4 2=63.4 3=86.3 Roosevelt Bridge 1=43.9 2=42.4 3=51.9 Dubois Park 1=47.1 2=31.5 3=38.0 Roosevelt Bridge

Dubois Park

Current Direction A (still, north, south, east, west)

Roosevelt Bridge

Dubois Park

Current Direction B (alongshore right to left, alongshore left to right, neither)

Roosevelt Bridge

Current Strength (strong, moderate, weak)

Roosevelt Bridge

Dubois Park

Dubois Park

1=21.5 2=31.8 3=27.0 1=20.9 2=40.5 3=48.8 3=4.0 4=21.3 5=13.5 1=23.7 2=74.9 3=79.0 4=0.0 5=112.0 1=34.1 2=16.6 3=34.1 1=67.8 2=53.7 3=34.1 1=28.2 2=26.4 3=26.9 1=23.0 2=61.8 3=35.3

ENT R2

F

p-values

0.002

0.72

0.5, no significant difference

0.03

12.6

<0.0001, significant difference between 1-2, 1-3

0.002

0.62

p=0.5, no significant difference

0.007

2.72

p=0.07, no significant difference

0.005

1.38

p=0.3, no significant difference

0.02

5.33

p=0.005, significant difference between 1-2, 1-3

0.06

0.40

p=0.7, no significant difference

0.10

0.48

p=0.7, no significant difference

0.02

5.68

p=0.004, significant difference between 1-2

0.01

3.17

p=0.04, significant difference between 1-3

0.0001

0.03

p= 0.1 no significant difference

0.007

2.27

p=0.1 no significant difference

Table 4. Results from ANOVA analyses, comparing enterococci levels versus categorical variables at Roosevelt Bridge (St. Lucie River) and Dubois Park (Loxahatchee River), 20002018

4

Canal Point Dependent Variables

Independent Variables Chlorophyll a (µg/L) Pheophytin (µg/L) Enterococci (CFU/100 mL) Nitrate-nitrite (mg/L) TKN (mg/L) Total Phosphorus (mg/L) Orthophosphorus (mg/L) Turbidity (NTU) Water Temperature (°°C) pH Salinity (ppt) DO (mg/L) Sample depth (m) Tide or wave height Roosevelt Bridge Dependent Variables

Independent Variables Chlorophyll a (µg/L) Pheophytin (µg/L) Enterococci (CFU/100 mL) Nitrate-nitrite (mg/L) TKN (mg/L) Orthophosphorus (mg/L) pH Salinity (ppt) Sample depth (m) Tide or wave height Dubois Park Dependent Variables

Independent Variables Chlorophyll a (µg/L) Pheophytin (µg/L) Enterococci (CFU/100 mL) Total Phosphorus (mg/L) Turbidity (NTU) Salinity (ppt) Sample depth (m)

Canal Point Canal Point Enterococci Chlorophyll a n=12, F= 1.20, n=12, F=3.94, p= 0.4, R2= 0.08 p=0.05, R2= 0.45 Parameter Estimate, p-value Not included Not included Not included Not included Not included Not included -1414, 0.1 Not included 131, 0.7 Not included Not included Not included -821, 0.8 Not included Not included Not included Not included Not included Not included 53.9, 0.08 1829, 0.4 -4.39, 0.99 208, 0.3 Not included Not included 126, 0.1 Not included Not included

Canal Point Pheophytin n=12 ,F= 9.79, p=0.04, R2=0.86 0.58, 0.03 Not included 0.008, 0.3 Not included Not included 36.9, 0.4 Not included Not included 3.08, 0.06 Not included -631, 0.1 15.7, 0.05 125, 0.2 2.66, 0.8

Roosevelt Roosevelt Enterococci Chlorophyll a n=8, F= 37.0, n=11, F= 63.9, 2 p= 0.1, R =0.97 p= 0.0006, R2= 0.99 Parameter Estimate, p-value 125, 0.3 Not included -87.8, 0.3 0.68, 0.0003 Not included 0.002, 0.3

Roosevelt Pheophytin n=8, F= 32.6, p= 0.03, R2=0.96

Not included Not included 1612, 0.1 915, 0.1 5.84, 0.5 -155, 0.8 Not included

-19.3, 0.4 Not included 49.6, 0.4 Not included -0.23, 0.5 Not included Not included

Not included 3.79, 0.2 Not included 1.92, 0.3 0.12, 0.4 Not included -0.37, 0.5

0.93, 0.2 Not included -0.001, 0.8

Dubois Dubois Enterococci Chlorophyll a n=12, F=3.79, n=12, F= 4.83, 2 p=0.06, R =0.43 p=0.04, R2 0.41 Parameter Estimate, p-value Not included Not included Not included Not included Not included Not included

Dubois Pheophytin n=12, F= 2.28, p= 0.16, R2= 0.19

-1641, 0.8 Not included 5.14, 0.7 225, 0.1

Not included 0.09, 0.1 -0.02, 0.8 Not included

41.5, 0.7 Not included -0.25, 0.4 Not included

Not included Not included Not included

Table 5. Measures significantly correlating with chlorophyll a, pheophytin, and enterococci, indicated by multiple regression, by monthly field sampling site. Not included=insufficient data to conduct multiple regression or comparison between dependent variable and itself omitted.

5

Rainfall Station (Water Source)

Station for Chlorophyll a, Pheophytin, or Enterococci (Water Source) Rainfall 24 Chlorophyll a Hours Before Pheophytin Enterococci Rainfall 3 Chlorophyll a Days Before Pheophytin Enterococci Rainfall Chlorophyll a Week Before Pheophytin Enterococci Rainfall Chlorophyll a Month Before Pheophytin Enterococci

Station USC00081276 at Canal Point (Lake Okeechobee) Canal Point (Lake Okeechobee) NS* NS 0.004 NS NS 0.09 NS 0.006 0.12 0.40 0.05 0.09

Station US1FLMT0016 at S308 (Lake Okeechobee) Canal Point (Lake Okeechobee) NS NS NS NS NS NS 0.01 0.03 0.11 0.36 0.31 0.12

Station US1FLMT0016 at S308 (Lake Okeechobee) Roosevelt Bridge (St. Lucie River)

Station USC00088620 at Roosevelt Bridge (St. Lucie River)

NS NS NS NS NS NS NS NS 0.04 0.04 0.02 0.45

NS NS NS NS NS NS NS NS NS NS NS 0.23

Roosevelt Bridge (St. Lucie River)

Station USC00084461 at Dubois Park (Loxahatchee River) Dubois Park (Loxahatchee River) NS 0.01 0.37 NS NS 0.29 NS 0.33 0.18 NS NS 0.05

Table 6. Regression R2 for significant (p≤0.05) chlorophyll a, enterococci, and pheophytin versus rainfall (NCEI) at Canal Point, Roosevelt Bridge, and Dubois Park 2017-2018. For S308 rainfall, chlorophyll a and enterococci from both Canal Point and Roosevelt Bridge were compared, as denoted. All r values are positive; *NS denotes a nonsignificant value.

6

Water Control Structure Water Quality Monitoring Site Monthly Sampling Site for Year-Long Study FDOH Monitoring Station

0

6km

Figure 1. General location of sampling sites and water control structures. Sampling sites include Canal Point, on Lake Okeechobee; Roosevelt Bridge, on the St. Lucie River; and Dubois Park, on the Loxahatchee River. Water control structures include S352, near Canal Point, and S308, approximately 13 km north, on Lake Okeechobee; S80, where the St. Lucie Canal connects to the St. Lucie River; and SE03, on the St. Lucie River. Basemap courtesy of U.S Census-TIGERweb.

1

Canal Point, Lake Okeechobee 35

a)

Rainfall (cm)

30 25 20 15 10 5

160

9-25-17 10-16-17 11-15-17 12-11-17

Chlorophyll

1-8-18

2-6-18

3-13-18

Pheophytin

4-10-18

5-15-18

6-12-18

7-10-18

Enterococci

8-14-18

3500

b)

3000

140

2500

120 100

2000

80

1500

60

1000

40

500

20

0

0 7

Enterococci (CFU/100 mL)

180

9-25-17 10-16-17 11-15-17 12-11-17

TN

1-8-18

TKN

2-6-18

3-13-18

Nitrate-Nitrite

4-10-18

5-15-18

Phosphorus

6-12-18

7-10-18

8-14-18

0.60

Orthophosphorus

6

0.50

c)

5

0.40

4

0.30 3

0.20

2

0.10

1

0.00

0 9-25-17 10-16-17 11-15-17 12-11-17

1-8-18

2-6-18

3-13-18

4-10-18

5-15-18

6-12-18

7-10-18

8-14-18

Sampling Date (MM/DD/YY)

Figure 2. a. Canal Point rainfall (cm) in the month before sampling (NCEI). b. Canal Point chlorophyll a, pheophytin, and enterococci. c. Canal Point TN, TKN, nitrate-nitrite, phosphorus, orthophosphorus

2

Phosphorus, Orthophosphorus (mg/L)

Nitrite-Nitrate, TN, TKN (mg/L)

Chlorophyll a and Pheophytin (µg/L)

0

Roosevelt Bridge, St. Lucie River Flow S308 (m3)

4000

a)

3000 2000 1000 0 80

9-25-17 10-16-17 11-15-17 12-11-17 1-8-18

2-6-18

3-13-18 4-10-18 5-15-18 6-12-18 7-10-18 8-14-18

40 20 0 45

9-25-17 10-16-17 11-15-17 12-11-17

40

Pheophytin ug/L

1-8-18

2-6-18

3-13-18 4-10-18

5-15-18

Chlorophyll

6-12-18

7-10-18

Enterococci

8-14-18

2500

c) 2000

35 30

1500

25 20

1000

15 10

500

5

0

0 9-25-17 10-16-17 11-15-17 12-11-17

1-8-18

2-6-18

3-13-18

4-10-18

5-15-18

6-12-18

7-10-18

8-14-18

Figure 3. a. Flow out of Lake Okeechobee (cubic meters per second) into the St. Lucie Canal during the monthly sampling period. b. Roosevelt Bridge rainfall (cm) in the month before sampling (NCEI). c. chlorophyll a, pheophytin, and enterococci. d. Roosevelt Bridge Roosevelt Bridge TN, TKN, nitrate-nitrite, orthophosphorus,

3

Enterococci (CFU/100mL)

Chlorophyll a and Pheophytin (µg/L)

Rainfall (cm)

b) 60

9-25-17

10-16-17 11-15-17 12-11-17

Chlorophyll

1-8-18

2-6-18

3-13-18

Pheophytin

4-10-18

5-15-18

6-12-18

Enterococci

7-10-18

12

8-14-18

700

b)

600

10

500

8

400

6

300

4

200

2

100

0 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

9-25-17

10-16-17 11-15-17 12-11-17

TN

TKN

1-8-18

2-6-18

Nitrate-Nitrite

3-13-18

4-10-18

5-15-18

Phosphorus

6-12-18

7-10-18

8-14-18

Orthophosphorus

0 0.06 0.05

c) 0.04 0.03 0.02 0.01 0.00 9-25-17

10-16-17 11-15-17 12-11-17

1-8-18

2-6-18

3-13-18

4-10-18

5-15-18

6-12-18

7-10-18

8-14-18

Sampling Date (MM/DD/YY)

Figure 4. a. Dubois Park Rainfall (cm) b. Dubois Park chlorophyll a, pheophytin, enterococci. c. Dubois Park TN, TKN, nitrate-nitrite, phosphorus, orthophosphorus. Flow data not available for Dubois Park.

4

Enterococci (CFU/100mL)

a)

Phosphorus, Orthophosphorus (mg/L)

Rainfall (cm) Chlorophyll, Pheophytin (µg/L) Nitrate-Nitrite, TKN, TN (mg/L)

70 60 50 40 30 20 10 0 14

Dubois Park, Loxahatchee River

6

4

Chlorophyll a (µg/L)

3 4

2.5

3

2 1.5

2

1 1

Enterococci (nromalized)

3.5

5

0.5

0

0 0

12

12.1

24

24.1

36

36.1

Hours

Chlorophyll a Enterococci

Algae Culture Room Incubator Symbol Water Sediment Symbol Water Sediment Lake Lake Lake Lake Okeechobee Okeechobee Okeechobee Okeechobee Lake Lake Okeechobee Okeechobee

Lake Lake Okeechobee Okeechobee

Figure 5. Microcosm experiment 3: Water and sediment from Lake Okeechobee, chlorophyll a and enterococci measured every 12 hours.

5

4

2

1.6

Chlorophyll (µg/L)

3 1.4 2.5

1.2

2

1 0.8

1.5

0.6 1 0.4 0.5

Enterococci (normalized)

1.8

3.5

0.2

0

0 0

12

12.1

24

24.1

36

36.1

Hours

Chlorophyll a Enterococci

Algae Culture Room Symbol Water Lake Okeechobee Lake Okeechobee

Sediment Dubois Park Dubois Park

Symbol

Water Lake Okeechobee Lake Okeechobee

Sediment Lake Okeechobee Lake Okeechobee

Figure 6. Microcosm experiment 4: Water and sediment from Lake Okeechobee and sediment from Dubois Park, chlorophyll a and enterococci measured every 12 hours.

6

Highlights • • • •

Microalgae blooms and enterococci exceedances have occurred simultaneously. Chlorophyll a and enterococci varied with nutrients at all sites. Different species of N/P associated with chlorophyll a and enterococci per site. Sediment may have contributed to blooms and exceedances during this study.

Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: