Draft - University of Toronto T-SpaceDraft Casas et al… 1 1 Title: Effect of temperature on chlorine treatment for elimination of freshwater 2 phytoplankton in ballast water: bench

Draft

Effect of temperature on chlorine treatment for elimination of freshwater phytoplankton in ballast water: bench scale

test

Journal: Canadian Journal of Fisheries and Aquatic Sciences

Manuscript ID cjfas-2018-0179.R2

Manuscript Type: Article

Date Submitted by the Author: 12-Nov-2018

Complete List of Authors: Casas-Monroy, Oscar; Fisheries and Oceans Canada, Great Lakes Laboratory for Fisheries and Aquatic SciencesVanden Byllaardt, Julie; Hamilton Harbour Remedial Action Plan OfficeBradie, Johanna; Fisheries and Oceans Canada, Great Lakes Laboratory for Fisheries and Aquatic SciencesSneekes, Andrea; Wageningen Marine ResearchKaag, Klaas; Wageningen Marine ResearchBailey, Sarah; Department of Fisheries and Oceans

Keyword: FRESHWATER < Environment/Habitat, INVASIVE SPECIES < Organisms, PHYTOPLANKTON < Organisms, TEMPERATURE EFFECTS < General

Is the invited manuscript for consideration in a Special

Issue? :Not applicable (regular submission)

https://mc06.manuscriptcentral.com/cjfas-pubs

Canadian Journal of Fisheries and Aquatic Sciences

Draft

Casas et al… 1

1 Title: Effect of temperature on chlorine treatment for elimination of freshwater

2 phytoplankton in ballast water: bench scale test

3

4 Oscar Casas-Monroy1*, Julie Vanden Byllaardt2, Johanna Bradie1, Andrea Sneekes3, Klaas

5 Kaag3, and Sarah A. Bailey1

6

7 1Great Lakes Laboratory for Fisheries and Aquatic Sciences, Fisheries and Oceans Canada,

8 Lakeshore Road, Burlington, ON, L7S 1A1, Canada

9 [email protected], [email protected]

10

11 2Hamilton Harbour Remedial Action Plan Office, Canada Centre for Inland Waters, 867

12 Lakeshore Road, Burlington, ON, L7S 1A1, Canada

13 [email protected]

14

15 3Wageningen Marine Research. PO Box 68 1780AB, Ijmuiden, the Netherlands


17

18 4Wageningen Marine Research. PO Box 57 1780A, Den Helder, the Netherlands


20

21 *Corresponding author: [email protected]

22 Phone number: 1-(905)-336-4728; FAX: 1-(905)-336-6437

Page 1 of 40



mailto:[email protected]






Draft

Casas et al… 2

23 ABSTRACT

24 Chlorine efficacy as a biocide for ballast water treatment was investigated under warm-cold

25 temperatures, across winter and summer seasons. Freshwater phytoplankton samples were

26 collected and acclimated under in situ environmental conditions ranging from 2-22 °C. Samples

27 were exposed to seven chlorine treatments (from 0.02 to 5.0 ppm), in addition to a control

28 (0.0 ppm). Free chlorine concentrations, phytoplankton abundance and photosynthetic efficiency

29 were measured up to 48 h following treatment. After 4 h of treatment at concentrations less than

30 0.2 ppm, phytoplankton densities were reduced more than 50 % without cell resurgence. Similar

31 reduction was recorded immediately after exposure when chlorine concentrations were higher

32 than 3 ppm. After 8 h, free chlorine neared 0.0 ppm for initial chlorine concentrations below

33 1.2 ppm irrespective of temperature regime. Winter phytoplankton exhibited slightly lower

34 mortality to chlorine exposure regardless of the temperature although they also exhibited lower

35 photosynthetic efficiency. Despite a general absence of significant effect of temperature on the

36 chlorine decay, our results suggest that higher doses of chlorine or longer exposure times may be

37 required during winter to achieve full treatment effect. Tests at large scales are needed to further

38 confirm these findings.

39

40 Key words: Ballast water management system, Invasive species, Treatment efficacy, PAM

41 fluorometry (3 to 8 words)

42

Page 2 of 40



Draft

Casas et al… 3

43

44 INTRODUCTION

45 Commercial shipping is one of the primary vectors for the transfer of harmful and non-

46 indigenous species (NIS) to bays and estuaries worldwide (Hulme 2009). Vessels not fully loaded

47 with cargo may need to fill their ballast tanks with water (including resident organisms) from a

48 port to enhance the ship's stability prior to a voyage. Once the ballast water is discharged during

49 cargo loading operations at the destination port, organisms that survived the voyage (e.g.,

50 darkness, decrease in oxygen and/or predation within ballast tanks) are released and have the

51 potential to establish new populations, posing serious ecological, economic and health threats to

52 the receiving regions (Roy et al. 2012).

53 In order to reduce and prevent introductions of harmful aquatic organisms and pathogens

54 in new regions, the International Maritime Organization (IMO) adopted the International

55 Convention for the Control and Management of Ship’s Ballast Water and Sediments (hereafter

56 the Convention) (IMO 2004), which entered into force in September 2017. The Convention sets a

57 timeline for ship owners and operators to comply with a new regulation (D-2) that sets a

58 maximum number of viable organisms in discharged ballast water, by size, including < 10 viable

59 organisms m-3 ≥ 50 µm in minimum dimension; < 10 viable organisms mL-1 ≥ 10 and < 50 µm in

60 minimum dimension; < 250 colony forming units (cfu) 100 mL-1 of Escherichia coli; < 100 cfu

61 100 mL-1 of Enterococci; and, < 1 cfu 100 mL-1 Vibrio cholerae (IMO 2004). The IMO D-2

62 regulation applies to ships engaged in international voyages. To comply with the IMO D-2

63 regulation, many ship owners will choose to install an on-board Ballast Water Management

64 System (BWMS) that is appropriate for the range of environmental conditions encountered

65 during a ship’s typical trade routes. Important factors considered are costs to purchase, install and

66 operate the BWMS and holding time (i.e., the minimum time required to achieve full treatment

Page 3 of 40



Draft

Casas et al… 4

67 efficacy), since some ships need to discharge their ballast water less than one day after uptake on

68 short voyages.

69 Presently, there are more than 50 BWMS on the market using different processes (i.e.,

70 mechanical, physical, and chemical) to remove organisms from ballast water (IMO 2017). Most

71 BWMS apply a primary filtration step (mechanical process) to remove organisms larger than 55

72 µm (e.g., eggs, larvae, and invertebrates) (Veldhuis et al. 2006). The secondary step involves the

73 use of a disinfection process (physical or chemical) such as UV irradiation, ozone, or chlorination

74 (Gregg et al. 2009; Tsolaki and Diamadopoulos 2010) to remove organisms smaller than 55 µm

75 (e.g., protists, cysts, bacteria). Depending on the BWMS technology, treatment efficacy may be

76 variable across taxonomic groups, for example, being more effective at killing smaller

77 phytoplankton cells than larger zooplankton or resistant life stages (Gregg et al. 2009; Lloyd's

78 Register Marine 2015). Conversely, different organism life stages may be more likely to tolerate

79 treatment than others (e.g., dinoflagellates and diatoms in encysted form). BWMS may also have

80 variable efficacy depending on ambient environmental conditions (e.g., sea temperature,

81 turbidity) during the application of treatment, influencing the risk of introduction and

82 establishment of organisms surviving treatment and having potential for population growth after

83 discharge (Grob and Pollet 2016).

84 Chemical disinfection using strong oxidants is one of the most commonly applied ballast

85 water treatments to prevent the introduction and establishment of harmful aquatic organisms and

86 pathogens (Lloyd’s Register Marine 2015). Among different oxidants, chlorine is a practical

87 biocide used to eliminate a broad range of organisms (e.g. algae, invertebrates, fish) from ballast

88 water (Añasco et al. 2008), with minimal or no harm upon ecosystems or environment due to

89 rapid decay or application of neutralizing agents. Chlorine is cost-effective at both small and

90 large scales, and its applications include treatment of municipal wastewater, sanitation of

Page 4 of 40



Draft

Casas et al… 5

91 swimming pools and agricultural farms, control of biofouling in coastal power stations, and

92 household disinfection (Añasco et al. 2008; Gregg et al. 2009). In fresh water, the dominant

93 reactive forms of chlorine are either hypochlorous acid (HOCl) (most toxic chlorine compound)

94 or its associated hypochlorite anion (OCl-), which act as oxidizing agents known as free chlorine.

95 In marine and estuarine waters, the presence of bromide ions and the higher reactivity of bromine

96 leads to the formation of chlorine-produced oxidants. Free chlorine may damage cell walls and

97 disrupt biomolecules such as DNA, RNA, fatty acids, cholesterol and proteins (Denyer 1995;

98 Werschkun et al. 2012).

99 Previous studies have shown that phytoplankton can survive in the darkness of ballast

100 tanks for up to 33 days (Doblin et al. 2004; Kang et al. 2010; Klein at al. 2010), and that after

101 treatment, a variety of photosynthetic organisms can exhibit population growth within 4 to 20

102 days of being returned to benign conditions (Liebich et al. 2012; Martínez et al. 2013; Katayama

103 et al. 2015). It is therefore of special interest to examine the survival potential (or reproduction)

104 of phytoplankton after chlorine ballast water treatment across a range of temperatures.

105 Few studies have examined the effect of temperature on chlorine chemical-based BWMS.

106 Traditional, intensive methods such as microscopy can provide direct measurements (i.e., number

107 of living organisms mL-1) before/after treatment, while indicative methods such as fluorometry

108 can provide more rapid but indirect estimates of cell abundance. Chlorophyll a fluorescence

109 techniques are used widely to assess the abundance and physiological responses of cyanobacteria

110 and microalgae. Chlorophyll a is one of the primary molecules that absorbs sunlight during

111 photosynthesis, but a fraction of the energy is reemitted as fluorescence. The raw fluorescence

112 reemitted by cells can be converted to an estimated number of living organisms based on

113 empirical relationships (i.e., using a calibration curve) between raw fluorescence values and

114 organism concentrations derived from epifluorescence microscopy counts. Devices such as the

Page 5 of 40



Draft

Casas et al… 6

115 Pulse Amplitude Modulation (PAM) fluorometer deliver a series of light pulses to assess baseline

116 fluorescence under dark adaptation (F0) and maximal fluorescence (Fm) under saturating light

117 (Wright et al. 2015); the difference, Fm − F0 (variable fluorescence or Fv), provides an indication

118 of the total active chlorophyll in a sample. These values also specify the quantum yield (e.g.,

119 Fv/Fm), or the photosynthetic efficiency of cells, where a value below 0.25 indicates that

120 phytoplankton cells are no longer photosynthetically efficient (Kolber and Falkowsky 1993;

121 Schreiber 1998; Welschmeyer 2014). When photosynthetic activity ceases, energy essential for

122 basic functioning is no longer produced and the organism dies (Veldhuis et al. 2001). This ratio

123 can be used in ballast water research in order to determine the health of organisms (Wright et al.

124 2015) as an indicator of the efficiency of BWMS.

125 Despite much progress during the past ten years, there is still a need to find effective and

126 environmentally sound strategies to manage and prevent the introduction of NIS and several

127 studies have indicated that BWMS cannot completely eliminate phytoplankton from ballast water

128 (see Gregg et al. 2009; Tsolaki and Diamadopoulos 2010). All BWMS need to get type approval

129 from the IMO, and from the U.S. Coast Guard for use in U.S. waters. Each BWMS is subjected

130 to Type Approval testing across a range of temperatures, salinities, species, and biomass, but this

131 does not guarantee reliable performance of BWMS across global environmental conditions.

132 There are few studies evaluating the effect of temperature on disinfection processes for

133 freshwater ballast, particularly at extreme temperatures (de Lafontaine et al. 2008a; Maranda et

134 al. 2013; Paolucci et al. 2015). As both international and domestic ships may use a chlorine-based

135 BWMS when loading ballast water at a variety of freshwater ports (such as in the Great Lakes,

136 Antwerp, Hamburg, Rotterdam, etc.) during the cold winter season, the objectives of this study

137 were to examine the effect of temperature and season on 1) the decay rate of free chlorine after

138 application to fresh water; 2) the rate and percent mortality of natural freshwater phytoplankton

Page 6 of 40



Draft

Casas et al… 7

139 after chlorine treatment; and 3) the photosynthetic efficiency of natural freshwater phytoplankton

140 after chlorine treatment.

141

142 MATERIAL AND METHODS

143 Experimental Design

144 Experiments were conducted at the Great Lakes Laboratory for Fisheries and Aquatic

145 Sciences in Burlington, Ontario, Canada, during winter and summer seasons using natural

146 freshwater phytoplankton communities collected from Hamilton Harbour (Lake Ontario). Six

147 experiments were conducted in triplicate for each season, beginning on separate days, with

148 independent in situ environmental conditions (Table 1). Treatment consisted of seven free

149 chlorine concentrations (ranging 0.02 – 5.0 ppm), and a control (0.0 ppm). Free chlorine, as well

150 as phytoplankton abundance and photosynthetic efficiency, were measured before and up to 48

151 hours after initial chlorine application. For summer experiments, phytoplankton collected from

152 22 °C harbour water were exposed to free chlorine concentrations at 22 °C ± 0.5 and 2 °C ± 0.5,

153 whereas winter experiments collected phytoplankton from 2 °C harbour water and experiments

154 were conducted at 2 °C ± 0.5 and 18 °C ± 0.5. The range between low and high temperatures

155 during winter was reduced to avoid mortality of cells, since winter plankton is more sensitive to

156 extreme temperature changes (Rasconi et al. 2017).

157

158 Phytoplankton collection

159 A large shipping container was used to construct a shoreside testing laboratory to pump water

160 directly from the harbour to the laboratory. The fresh water was collected using a hose system

161 attached to a non-submersible pump (WEG Washdown-HP. 1.00(0.75)-D56C-RPM1750) and

162 pre-filtered with a 500 µm Nitex mesh in order to remove large debris. Prior to experiments, 5 L

Page 7 of 40



Draft

Casas et al… 8

163 of harbour water were also filtered on hydrophilic polysulfone 0.2 µm pore size membranes

164 (Mini Capsule, Pall Corporation) and stored at 4 °C; this water was used to rinse all the biological

165 material of the subsamples. Similarly, a continuous subsample of harbour water (at least 100 L)

166 was filtered through 50 µm and 10 µm (in diagonal) Nitex mesh and the filtrand (phytoplankton

167 cells) retained on the 10 µm mesh was rinsed into a 1 L Nalgene bottle using the 0.2 µm-filtered

168 harbour water. Harbour water temperature, oxygen and pH were measured with a YSI 556

169 (Yellow Springs Instruments, Ohio, USA) probe during collection in order to account for any

170 external effect on the experiment’s results. In particular, we confirmed that pH measured between

171 5 and 7 to ensure the production of the most effective compounds of free chlorine (hypochlorous

172 acid – HOCl) during the experiments (Delacroix et al. 2013). Enumeration and taxonomic

173 identification of collected phytoplankton cells was conducted as part of a simultaneous study (see

174 Casas-Monroy et al. 2017). Briefly, 100 mL of sample water were preserved with Lugol's Acid

175 and a total volume of 50 mL was put in a settling column for 24 hours following the Utermöhl

176 technique (Utermöhl, 1958), and intact cells counted using a Nikon AZ100 inverted microscope.

177

178 Preparation of phytoplankton

179 Immediately after collection, the 1 L phytoplankton sample was acclimated in the dark for 24

180 hours using a Percival Scientific Inc. incubator (Model GR36L) under in situ environmental

181 conditions, or in the case of a test with a change in temperature; acclimation was progressive

182 (approx. 1 °C change per hour). A calibration curve was created for every experiment in order to

183 convert in vivo fluorescence values to the number of phytoplankton cells mL-1, accounting for

184 changes in dominant phytoplankton groups. In vivo fluorescence was measured using a calibrated

185 Turner model 10 AU fluorometer (Turner Designs, Sunnyvale, California, USA) equipped with

186 an in vivo chlorophyll filter and sample chamber. Living cells were enumerated in a Sedgwick-

Page 8 of 40



Draft

Casas et al… 9

187 Rafter chamber using fluorescein diacetate following Adams et al. (2014) under a Nikon AZ100

188 compound epifluorescence microscope using blue light excitation-green band pass emission filter

189 cubes (FITC) in combination with a DAPI-FITC-Texas Red filter cube. Following acclimation,

190 the 1 L phytoplankton sample was diluted to 5 L in a Nalgene carboy using filtered harbour water

191 (0.2 µm), and held at a test temperature (2 °C or 18/22 °C), with the desired minimum number of

192 cells (~700 to 1000 cells mL-1) verified via the calibration curve (Table A1).

193

194 Chlorine treatment preparation

195 To begin each test, the natural chlorine demand of Hamilton Harbour water was measured

196 in order to adjust experimental free chlorine concentrations as close as possible to the nominal

197 concentrations: 0 ppm (control), 0.02 ppm, 0.20 ppm, 0.70 ppm, 1.20 ppm, 2.20 ppm, 3.00 ppm,

198 5.00 ppm. Chlorine demand was determined by adding 7.6 % Clorox bleach to filtered harbour

199 water (0.2 µm) in a series of increasing amounts until the free chlorine concentration became

200 measurable via a Colorimeter (900 Chlorine Colorimeter, using 0.02 – 8.0 mg L-1 chlorine

201 powder; Hoskin Scientific Ltd.). Experimental free chlorine concentrations were then prepared

202 via a stock solution of chlorine (50 ppm using 7.6 % bleach) by adding the measured chlorine

203 demand plus the nominal concentration into a graduated cylinder containing 500 mL of the

204 phytoplankton sample. The mixtures were stirred thoroughly and poured into 60 mL borosilicate

205 tubes.

206

207 Approaches to measure variables

208 Free chlorine concentrations were measured from one tube per chlorine treatment at each

209 of eight time points (56 tubes total). Each tube was subsampled for three measurements using the

210 Colorimeter described above.

Page 9 of 40



Draft

Casas et al… 10

211 Three replicate tubes for each chlorine concentration (including the control) served to

212 track the toxic effects of chlorine on phytoplankton abundance by measuring the in vivo

213 fluorescence with a Turner 10 AU fluorometer (21 tubes total). Measurements were repeated on

214 the same tubes at every time point. Percentage mortality of phytoplankton (Table A1) was

215 calculated as follows:

216 100 ― ( 𝑎𝑏𝑢𝑛𝑑𝑎𝑛𝑐𝑒 [𝐴]𝑎𝑏𝑢𝑛𝑑𝑎𝑛𝑐𝑒[𝐶] ∗ 100)

217 Where A represents the mean abundance of cells mL-1 for a given free chlorine

218 concentration at time (t) and C represents the abundances of cells mL-1 in the controls at time (t).

219 The quantum yield, or photosynthetic efficiency (Fv/Fm) (Schreiber 1998; Kolber and

220 Falkowsky, 1993), was examined using one tube for each control, 0.22, 1.2 and 3.0 ppm chlorine

221 treatments and analyzed at each of the eight time points (32 tubes total). Each tube was

222 subsampled for three measurements with a BW680 Pocket Fluorometer (henceforth called the

223 Pulse Amplitude Modulation (PAM) fluorometer; Hach Company, Loveland, Colorado, USA)

224 following Casas-Monroy et al. (2016).

225 Between measurements and over the 48 hours of each test, tubes were incubated in an

226 environmental chamber controlled for temperature (i.e., 2 °C and 18 °C/22 °C ± 0.5 °C) and

227 photoperiod (i.e., 48 hours without light, to simulate ballast tank conditions).

228

Page 10 of 40



Draft

Casas et al… 11

229 Statistical analysis

230 All statistical analyses were performed in the R statistical programming environment (R

231 Core Team, 2018). The twelve tests (6 during summer and 6 during winter) were analyzed as

232 replicates. All variables measured (i.e., abundance, fluorescence, free chlorine concentration and

233 photosynthetic efficiency) were analyzed and averaged among readings. An analysis of variance

234 (ANOVA) was performed to determine if any differences existed in the decay of free chlorine

235 during the full 48 hours of the experiment among initial concentrations vs. free chlorine effective

236 concentrations, hours (at each sampling time point) or temperatures.

237 The effect of temperature on chlorine concentration and phytoplankton abundance were

238 analyzed by a linear mixed model, with fixed effects for treatment, free chlorine concentration,

239 temperature, season, exposure time, and their two-way interactions. Time and concentration were

240 log-transformed prior to analysis to linearize the relationship and phytoplankton abundance was

241 square-root-transformed to stabilize variance. We specified a random intercept for test and tube,

242 and allowed an uncorrelated random slope for time within test. Different variances per stratum

243 were specified for seasons, treated versus control, and temperature. The model was fit using

244 restricted maximum likelihood and backwards selection was conducted using likelihood ratio

245 tests so that only significant predictors remained in the final model. Visual inspection of model

246 residuals was completed for model validation.

247 Finally, we conducted a series of independent t-test analyses of two samples assuming

248 unequal variance between the photosynthetic efficiency (Fv/Fm) and the threshold of 0.25. The

249 effect of temperature on the photosynthetic efficiency (Fv/Fm) was assessed to test the null

250 hypothesis that Fv/Fm did not differ from a threshold of 0.25 (at each sampling time point,

251 concentration and temperature). The P-value was adjusted for multiple tests.

252

Page 11 of 40



Draft

Casas et al… 12

253 RESULTS

254 Free chlorine decay

255 Temperature did not have a strong effect on decay rates of free chlorine. Differences in

256 chlorine concentration between temperatures were observed only for the highest nominal chlorine

257 concentration (5.0 ppm) at 24 hours and 48 hours after dosing during the summer phytoplankton

258 testing (df = 256; SE = 0.12; P < 0.018) (Tables 2a and 2b), where higher free chlorine

259 concentrations were observed at the higher experimental temperature (22 °C). The free chlorine

260 concentration appeared to decay at a slower rate during the winter (Fig. 1) vs. summer testing

261 (Fig. 2), particularly at initial nominal concentrations of 1.2 ppm and greater (Figs. 1 and 2).

262 Some free chlorine (less than 0.5 ppm) remained after 48 hours during winter testing (Fig. 1),

263 while free chlorine concentration typically decreased to 0 ppm during summer testing, there was

264 significant chlorine still present at 22 °C in the 5 ppm group (Fig. 2).

265 Natural abundances of phytoplankton varied at time 0 of each test according to season

266 (i.e., initial phytoplankton abundance in harbour water collected in March and early April was

267 approximately twice that of harbour water collected in August) (Table 1). Identification of taxa

268 indicated that Bacillariophyceae was the dominant taxon during all experiments, ranging from

269 36 % to 86 % of total abundance. Winter community samples contained secondary taxa

270 Dinophyceae (8 %), Chlorophyceae (4 %) and Dictyochophyceae (2 %), while summer

271 community samples also contained Cyanophyceae (22 %) and Chlorophyceae (3 %). At the

272 species level, Asterionella formosa was dominant during the winter while Synedra sp. was

273 dominant during the summer.

274

275

276

Page 12 of 40



Draft

Casas et al… 13

277 Phytoplankton Mortality

278 Comparing controls and treated samples, our mixed-effect model showed that treated

279 samples had significantly lower abundances than control samples (β = -4.39, SE = 0.50,

280 P < 0.001) at all time points (0 to 48 hours), at both temperatures, for all tests. Up to 25 %

281 mortality was observed in controls with a significant decrease in phytoplankton abundance for

282 both control and treated samples across time (β = -3.13, SE = 0.21, P < 0.001), however the

283 decrease was three times higher for treated samples (β = -5.24, SE = 0.18, P < 0.001).

284 The rate of mortality was typically steeper at the lower experimental temperature, with

285 percent mortality being, on average, 18 % higher after 1 hour of exposure for tests conducted at

286 2 °C vs. 18/22 °C (Figs. 3 and 4). After 48 hours of exposure, however, the effect of temperature

287 had decreased, with less than 10 % difference in cumulative mortality across temperatures. A

288 similar but opposite effect of season was observed, with mortality being, on average, 17 % lower

289 after 1 hour of exposure during winter testing compared to summer testing. Again, the difference

290 decreased with time, with about 10 % difference between seasons after 48 hours of exposure.

291 Our mixed-effect model confirmed significance of these results, showing lower cell

292 mortality at 18 °C (β = -4.53, SE = 0.45, P < 0.001) during the winter tests, despite higher cell

293 concentrations particularly at low temperatures (β = 3.48, SE = 0.64, P < 0.001). Comparing

294 results across the seasonal tests, the mixed-effect model confirmed a greater mortality of

295 phytoplankton with higher treatment concentration (β = -1.30, SE = 0.11, P < 0.001), and an

296 interaction between exposure time and treatment concentration with a greater decrease across

297 time at higher treatment concentrations (β = -0.60, SE = 0.05, P < 0.001) (Table 3). The increased

298 efficacy of higher concentrations of chlorine treatment was slightly reduced in the winter at low

299 temperatures (β = 0.27, SE = 0.14, P < 0.05).

Page 13 of 40



Draft

Casas et al… 14

300

301 Photosynthetic efficiency

302 Measurements of photosynthetic efficiency were typically higher at the lower

303 experimental temperature but lower during the winter testing. During winter tests, photosynthetic

304 efficiency measurements of phytoplankton maintained at 2 °C were significantly higher (t-test,

305 P < 0.001) than the threshold of 0.25 during the first 2 hours of treatment. After this period, the

306 Fv/Fm was either not significantly different or lower than the threshold of 0.25. For winter tests

307 conducted at 18 °C, photosynthetic efficiency measurements declined sooner, after only one hour

308 of treatment (Fig. 5).

309 During summer tests, photosynthetic efficiency measurements of phytoplankton

310 maintained at 2 °C were significantly higher than the 0.25 threshold across the 48 hours of the

311 experiment for all free chlorine concentrations (t-test, P < 0.001) (Fig. 6). For those cells

312 maintained at 22 °C, photosynthetic efficiency was significantly higher than 0.25 after 4 hours,

313 but after 8 hours of treatment, only few Fv/Fm values were recorded above the threshold (Fig. 6).

314 Control treatments showed high photosynthetic efficiency (Fv/Fm 0.50 – 0.75) throughout

315 the experiment (Fig. 5 and Fig. 6) (t-test, P < 0.0001).

316

317 Discussion

318 Our study found that phytoplankton cells collected during the winter and treated with

319 chlorine had slightly lower rates of mortality than phytoplankton cells collected during the

320 summer and treated with the same chlorine concentrations, despite a general absence of any

321 effect of temperature on the decay of chlorine. As cell activity was very low after chlorine

322 exposure during winter testing (as indicated by low Fv/Fm values) while the few cells surviving

323 exposure during summer testing showed moderate activity after 48 hours (Fv/Fm values reduced

Page 14 of 40



Draft

Casas et al… 15

324 but remain above the threshold value), these results suggest that the timeframe to achieve full

325 toxic effect of chlorine may be extended when applied during the winter season. Interestingly,

326 higher phytoplankton mortality was observed under free chlorine exposure at lower temperatures

327 for both cells collected during the winter and summer. In all, our results suggest that chlorine

328 application may be most effective when applied at ambient temperature, and that higher doses or

329 longer exposure times may be required during winter to achieve full treatment effect.

330 The degradation of chlorine was more rapid at higher free chlorine concentrations during

331 summer tests compared to winter tests, but we found no consistent effect of temperature during

332 this study. The degradation of chlorine, particularly for nominal concentrations lower than

333 1.2 ppm, was similar at both temperatures. Free chlorine concentrations at 2.2 and 3 ppm showed

334 a slower decay curve during winter compared to summer (regardless of temperature), while free

335 chlorine concentration was significantly different between both temperatures only at the highest

336 chlorine concentration during summer testing. These results indicate that chlorine may not be

337 utilized as quickly when applied in the winter, although this may not be a direct effect of

338 temperature. Few studies have considered temperature as an important variable affecting BWMS

339 efficacy (Drillet et al. 2012), although studies examining the effectiveness and residual toxicity of

340 active substances such as PERACLEAN® Ocean (Gregg and Hallegraeff 2007; de Lafontaine et

341 al 2008a, 2008b), chlorine dioxide (Maranda et al. 2013) and chlorine (Paolucci et al. 2015) have

342 noted the importance of temperature on reaction rates, where low temperature (and low pH) slow

343 down the chemical decomposition rate. De Lafontaine et al. (2008a) and Stehouwer et al. (2015)

344 also found Peraclean® Ocean to be more toxic at low temperatures for both freshwater and

345 marine phytoplankton. If, like other active substances, the degradation of chlorine is inversely

346 related to temperature (e.g., Peracetic acid; Kunigk et al. 2001), it will take longer to achieve

347 complete mortality when used in cold, fresh waters.

Page 15 of 40



Draft

Casas et al… 16

348 Data collected during a parallel study showed that UVT values (UV transmittance) for

349 Hamilton Harbour water ranged between 80.4 % and 84.6 % during summer and between 88.0 %

350 and 89.6 % during winter (see Casas-Monroy et al. 2017.). UVT measures the amount of light

351 that passes through a natural water sample compared to that which passes through a pure water

352 sample (i.e., distilled water). UVT is related to the quantity of organics, colloidal solids and other

353 material in the water, which absorb and scatter the light as it passes through the water column

354 (RealTech Inc., per. com.). Water with lower UVT will have higher baseline chlorine demand,

355 possibly contributing to differences observed in chlorine decay across seasons, however, the

356 prolonged availability of free chlorine during winter did not result in higher mortality. As

357 chlorine concentration approached zero after 48 hours of dosing irrespective of temperature or

358 season, the observed differences in decay rate will be most informative for treatment applications

359 by ships undertaking short voyages.

360 The reduction in phytoplankton abundance between control samples and treated samples

361 was significantly different, demonstrating a strong biocidal effect of chlorine within 48 hours.

362 The effectiveness of chlorine as a biocide for bacterial and phytoplankton communities is well

363 established (e.g., Gregg and Hallegraeff 2007; Maranda et al. 2013), with high effectiveness at

364 concentrations ranging from 4 to 50 ppm. Despite this, during our experiments, the winter

365 phytoplankton community showed lower mortality at both temperatures than the summer

366 phytoplankton community. Cell survival could be explained as an ability to escape the initial

367 biocide assault rather than as a response to temperature (Maranda et al. 2013), however,

368 temperature can influence the behavior and functionality of organisms, modifying their

369 metabolism rates, making them more or less resistant to treatments (Winder and Sommer 2012).

370 If there are inherent differences in the ability to survive chlorine treatment due to differences in

371 composition of winter vs. summer phytoplankton communities, this finding represents an

Page 16 of 40



Draft

Casas et al… 17

372 important knowledge gap – additional work is needed to confirm if phytoplankton originating

373 from Arctic or cold-temperate ecosystems during winter seasons may have higher tolerance than

374 the summer populations typically examined during type approval testing of BWMS. If yes, then

375 higher doses or longer exposure times may be needed to effectively manage ballast water under

376 these conditions.

377 The chemistry of chlorinated fresh water is very different to the chemistry of chlorinated

378 sea and brackish waters. In fresh water, applied chlorine will be hydrolysed into hypochlorous

379 acid (HOCl) and hypochlorite ion (OCl-). By introducing chlorine to seawater systems, a series of

380 redox-reactions take place, and several reactive intermediates are formed. Chlorine can rapidly

381 oxidize bromide ion (Br-) and iodide ion (I-) to form aqueous bromine (HOBr/OBr-) and aqueous

382 iodine (HOI/OI-), respectively (Westerhoff et al. 2004), which could result in slower degradation

383 of chlorine in freshwater ecosystems (de Lafontaine et al. 2008a). As such, it will be important to

384 ensure free chlorine is neutralized before discharging treated ballast water into the environment

385 (Bakalar 2016). The IMO has set maximum allowable discharge concentrations specific for each

386 BWMS using active substances, typically 0.2 ppm Total Residual Oxidants. Most BWMS using

387 active substances now measure TRO level during ballast water discharge and apply neutralizers

388 to the flow, when required.

389 To ensure maximum effectiveness and environmental safety of ballast water treatment

390 using chlorine, additional research may be needed to determine appropriate treatment

391 concentrations considering variable degradation rates and organisms tolerances under global

392 ballast water conditions. Chemical reactions and toxicity may change within rust-corroded

393 metallic tanks (de Lafontaine et al. 2008b) compared to the environmental conditions tested in

394 this paper. Despite a general absence of significant effect of temperature on the decay of chlorine,

395 our results suggest that higher doses of chlorine or longer exposure times may be required during

Page 17 of 40



Draft

Casas et al… 18

396 winter to achieve full treatment effect. Tests at large scales are needed to further confirm these

397 findings.

398

399 ACKNOWLEDGMENTS

400 This study was funded by Transport Canada and Fisheries and Oceans Canada (SAB). The

401 authors gratefully acknowledge the guidance and assistance given by Dr. Ève Gilroy

402 (Environment and Climate Change Canada) during various phases of this work. We also thank

403 Adam Van Der Lee for assistance with R, and Dawson Ogilvie and three anonymous reviewers

404 for constructive comments on earlier drafts.

405

406 REFERENCES

407 Adams, J.K., Briski E, Ram, J.L., Bailey, S.A. 2014. Evaluating the response of freshwater

408 organisms to vital staining. Manage Biol. Invasions 5(3): 197-208.

409 doi.org/10.3391/mbi.2014.5.3.02

410 Añasco, N., Koyama, J., Imai, S., Nakamura, K. 2008. Toxicity of residual Chlorines from

411 Hypochlorite-treated seawater to marine amphipod Hyale barbicornis and Estuarine Fish

412 Oryzias javanicus. Water Air Soil Pollut. 195: 129-136. doi.10.1007/s11270-008-9732-x

413 Bakalar, G. 2016. Comparisons of interdisciplinary ballast water treatment systems and

414 operational experiences from ships. SpringerPlus 5:240-252. Doi.10.1186/s40064-016-1916-z

415 Connel, G.F. 2006. Key operating strategies for chlorine disinfection operating systems.

416 Proceedings of the Water Environment Federation. Dallas, Texas. 61p.

417 Casas-Monroy, O., Chan, P.S., Linley, R.D., Vanden Byllaardt, J., Kydd, J., Bailey, S.A. 2016.

418 Comparison of three techniques to evaluate the number of viable phytoplankton cells in ballast

Page 18 of 40



Draft

Casas et al… 19

419 water after ultraviolet irradiation treatment. J. Appl. Phycol. 28(5) 2821-2830. doi

420 10.1007/s10811-016-0798-3

421 Delacroix, S., Vogelsang, C., Tobiesen, A., Liltved, H. 2013. Disinfection by-products and

422 ecotoxicology of ballast after oxidative treatment – results and experiences from seven years

423 of full-scale testing of ballast water management systems. Mar. Poll. Bull. 73(1): 24-36.

424 Doi.org/10.1016-j.marpolbul.2013.06.014

425 de Lafontaine, Y., Despatie, S-P., Wiley, C. 2008a. Effectiveness and potential toxicological

426 impact of the PERACLEAN Ocean ballast water treatment technology. Ecotox. Environ.

427 Safety 71(2): 355-369. doi:10.1016/j.ecoenv.2007.10.033

428 de Lafontaine, Y., Despatie, S-P., Veilleux, E., Wiley, C. 2008b. Onboard ship evaluation of the

429 effectiveness and the potential environmental effects of PERACLEAN Ocean for ballast water

430 treatment in very cold temperatures. Environ. Toxicol. 24(1): 49-65. doi10.1002/tox.20394

431 Denyer, S.P. 1995. Mechanisms of action of antibacterial biocides. International Biodeterioration

432 & Biodegradation. 36(3): 227-245

433 Drillet, G., Schmoker, C., Trottet, A., Mahjoub, M., Duchemin, M., Andersen, M. 2012. Effects

434 of temperature on type approval testing of ballast water treatment systems. Integr. Environ.

435 Assess. Manag. 9(2): 192-195. doi:10.1002/ieam.1394

436 Gregg, M., Rigby, G., Hallegraeff, G.M. 2009. Review of two decades of progress in the

437 development of management options for reducing or eradicating phytoplankton, zooplankton

438 and bacteria in ship’s ballast water. Aquat. Invasions 4(3): 521-565.

439 doi:10.3391/ai.2009.4.3.14

440 Gregg, M., Hallegraeff, G.M. 2007. Efficacy of three commercially available ballast water

441 biocides against vegetative microalgae, dinoflagellate cysts and bacteria. Harmful Algae 6(4):

442 567-584. doi:10.1016/j.hal.2006.08.009

Page 19 of 40



Draft

Casas et al… 20

443 Grob, C., Pollet, B.G. 2016. Regrowth in ships ballast water tanks: think again! Mar. Pol. Bul.

444 109(1): 46-48. doi.org/10.1016.j.marpolbul.2016.04.061

445 Hulme, P.E. 2009. Trade, transport and trouble: managing invasive species pathways in an era of

446 globalization. J. Appl. Ecol. 46: 10-18. doi: 10.1111/j.1365-2664.2008.01600.x

447 IMO (International Maritime Organization). 2004. International Convention for the Control

448 and Management of Ships' Ballast Water and Sediments. International Maritime

449 Organization (Adopted on 13 February 2004, in London, England; entry into force 8

450 September 2017;

451 http://www.imo.org/en/About/Conventions/ListOfConventions/Pages/International-

452 Convention-forthe-Control-and-Management-of-Ships'-Ballast-Water-and-Sediments-

453 (BWM).aspx (accessed 11.04.18)).

454 International Maritime Organization (IMO 2017). List of Ballast Water Management Systems

455 that Make Use of Active Substances Which Received Basic and Final Approval. Available

456 online:

457 http://www.imo.org/en/OurWork/Environment/BallastWaterManagement/Documents/BWM.2

458 -CIRC.34-REV.6.pdf (accessed July 2018).

459 Kang, J.H., Hyun, B.G., Shin, K. 2010. Phytoplankton viability in ballast water from

460 international commercial ships berthed at ports in Korea. Mar. Poll. Bull. 60(2): 230-237.

461 doi:10.1016/j.marpolbul.2009.09.021

462 Kolber, Z., Falkowski, P.G. 1993. Use fo active fluorescence to estimate phytoplankton

463 photosynthesis in-situ. Limnol. Ceanol. 38(8): 1646-1665.

464 Kunigk, L., Gomes, D.R., Forte, F., Vidal, K.P., Gomes, L.F., Sousa, P.F. 2001. The influence of

465 temperature on the decomposition kinetics of peracetic acid in solutions. Braz. J. Chem. Eng.

466 18(2): 217–220

Page 20 of 40



http://www.imo.org/en/About/Conventions/ListOfConventions/Pages/International-Convention-forthe-Control-and-Management-of-Ships'-Ballast-Water-and-Sediments-(BWM).aspx



Draft

Casas et al… 21

467 Liebich, V., Stehouwer, P.P., Veldhuis, M. 2012. Re-growth of potential invasive phytoplankton

468 followingUV-based ballast water treatment. Aquat. Invasions 7(1):29–36.

469 doi:10.3391/ai.2012.7.1.004

470 Lloyd’s Register Marine. 2015. Ballast water treatment technologies and current system

471 availability. Understanding Ballast Water Management Series. www.globallast.imo.org⁄wp-

472 content⁄uploads⁄2015⁄01⁄BM-Treatment-Technology-Sept-2012.pdf (accessed 11 November

473 2016)

474 Maranda, L., Cox, A.M., Campbell, R.G., Smith, D.C. 2013. Chlorine dioxide as a treatment

475 water to control invasive species: shipboard testing. Mar. Poll. Bull. 75(1): 76-89.

476 doi.org/10.1016/j.marpolbul.2013.08.002

477 Martinez, L.F., Mahamud, M.M., Lavin, A.G., Bueno, J.L. 2013. The regrowth of phytoplankton

478 cultures after UV disinfection. Mar. Pollut. Bull. 67:152-157.

479 doi.org/10.1016/j.marpolbul.2012.11.019

480 Paolucci, E., Hernandez, M., Potapov, A., Lewis, M.A., MacIsaac, H.J. 2015. Hybrid system

481 increase efficiency of ballast water treatment. J. Appl. Ecol. 52(2): 348-357.

482 doi:10.1111/1365-2664

483 Rasconi, S., Winter, K., Kainz, M.J. 2017. Temperature increase and fluctuation induce

484 phytoplankton biodiversity loss – Evidence from a multi-seasonal mesocosm experiment.

485 Ecol. Evol. doi: 10.10002/ece3.2889

486 R Core Team. 2018. R: A language and environment for statistical computing. R Foundation for

487 statistical Computing. Vienna, Austria. Available at: http://www.R-project.org/

488 Roy, S., Parenteau, M., Casas-Monroy, O., Rochon, A. 2012. Coastal ship traffic: a significant

489 introduction vector for potentially harmful dinoflagellates in eastern Canada. Can. J. Fish.

490 Aquat. 69: 627-644. doi:10.1139/F2012-008

Page 21 of 40



http://www.R-project.org/

Draft

Casas et al… 22

491 Schreiber, P. 1998 Chlorophyll fluorescence: new instruments for special applications. In:

492 Garabb G. (ed) Photosynthesis: Mechanisms and effects. Kluwer Academic Publishers,

493 Dordrecht Vol V. pp 4253-4258.

494 Stehouwer, P.P., Bruma, A., Peperzak, L. 2015. A comparison of six different ballast water

495 treatment systems based on UV radiation, electrochlorination and chlorine dioxide.

496 doi:10.1080/09593330.2015.1021858

497 Tsolaki, E., Diamadopoulos, E. 2010. Technologies for ballast water treatment: a review, J.

498 Chem. Technol. Biotechnol. 85: 19-32. doi:10.1002/jctb.2276

499 Utermöhl, H., 1958. Zur Vervollkommnung der Quantitativen Phytoplankton-Methodik.

500 Mitt. Int. Ver. Theor. Angew. Limnol. 9: 1–38.

501 Veldhuis, M., Kraay, G., Timmermans, K. 2001. Cell death in phytoplankton: correlation

502 between changes in membrane permeability photosynthetic activity, pigmentation and growth.

503 European J. of Phycol. 36: 167-177

504 Veldhuis, M., Fuhr. F., Boon, J.P., Ten Hallers-Tjabbers, C.C. 2006. Treatment of ballast

505 water; how to test a system with a modular concept? Environ. Technol. 27:909–921.

506 Welschmeyer, N., 2014. Variable chlorophyll a fluorescence as a rapid ballast water compliance

507 test. Standard Operating Procedure (Annex 2, Appendix 1) in IMO Document

508 MEPC66/INF.27 Ballast Water Sampling Methods for Assessing Compliance With the

509 Standards of the International Convention for the Control and Management of Ships' Ballast

510 Water and Sediments (Submitted by SGS S.A., Environmental Services, Geneva,

511 Switzerland; SGS Institut Fresenius GmbH, Taunusstein, Germany; and Dr. Nick

512 Welschmeyer, Moss Landings Marine Laboratories, California, United States. 33–38 pp.).

Page 22 of 40



Draft

Casas et al… 23

513 Werschkun, B., Sommer, Y., Banerji, S. 2012. Disinfection by-products in ballast water

514 treatment: an evaluation of regulatory data. Water Res. 46: 4884-4901.

515 doi:10.1016/j.watres.2012.05034

516 Westerhoff, P., Chao, P., Mash, H. 2004. Reactivity of natural organic matter with aqueous

517 chlorine and bromine. Water Res. 38: 1502-1513.

518 Winder, M., Sommer, U. 2012. Phytoplankton response to a changing climate. Hydrobiologia

519 698: 5-16. doi:10.1007/s10750-012-1149-2

520 Wright, D.A., Welschmeyer, N.A., Peperzak, L., 2015. Alternative, indirect measures of ballast

521 water treatment efficacy during a shipboard trial: a case study. J. Mar. Eng. Technol. 14:1–8.

522 doi.org/10.1080/20464177.2015.1022379

523

Page 23 of 40



Draft

Casas et al… 24

524 Table 1. In situ environmental conditions of Hamilton Harbour water. DO = Dissolved oxygen.

Test Sampling DatesInitial

abundances Salinity Temperature pH DO DO(Cell mL-1) (psu) (°C) (%) (mg/L)

I 25-Mar 1278 0.22 1.69 5.63 111.1 15.47II 01-Apr 1249 0.32 2.94 5.59 147.5 19.84III 08-Apr 1259 0.41 4.33 4.58 131 16.98IV 13-Apr 696 0.38 4.07 4.92 134.1 17.5V 15-Apr 709 0.34 6.44 5.58 127.8 16.11VI 21-Apr 698 0.4 7.53 5.24 128.8 15.39VII 22-Jul 452 0.34 23.49 6.21 123.2 10.46VIII 27-Jul 451 0.35 23.14 6.24 140.8 12.02IX 30-Jul 450 0.33 25.89 6.29 159.4 12.93X 05-Aug 531 0.33 22.91 7.04 125.1 10.73XI 10-Aug 538 0.33 22.35 7.05 141 12.22XII 12-Aug 532 0.32 20.52 7.11 99.4 10.52

525

526

Page 24 of 40



Draft

Casas et al… 25

527

528 Table 2. Analysis of variance for free chlorine concentrations for A) winter phytoplankton

529 population kept at 2 °C and 18 °C; B) for summer phytoplankton population kept at 22 °C and

530 2 °C.

A) Source Df Sum Sq Mean Sq F value Pr (<F)Concentration (ppm) 7 40.74 5.8194 88.7827 0.0001Temperature (°C) 1 0.044 0.0435 0.6639 0.4160Hour 7 7.732 1.1046 16.8527 0.0001Concen:Temperature 7 0.393 0.0561 0.8563 0.5418Concen:hour 49 3.0332 0.0619 0.9441 0.5825Temperature:hour 7 0.240 0.0343 0.5235 0.8165Concen:hour:Temperature 49 0.493 0.0101 0.1535 1.0000 Residuals 256 16.780 0.0655

531

B) Source Df Sum Sq Mean Sq F value Pr (<F)Concentration (ppm) 7 49.1 7.0087 300.7555 0.0001Temperature (°C) 1 0.73 0.7318 31.4025 0.0001Hour 7 16.5 2.3556 101.0815 0.0001Concen:Temperature 7 0.61 0.0870 3.7328 0.0001Concen:hour 49 8.2 0.1680 7.2079 0.0001Temperature:hour 7 0.21 0.0296 1.2684 0.266Concen:hour:Temperature 49 1.8 0.0372 1.5978 0.01Residuals 256 5.97 0.233

532

Page 25 of 40



Draft

Casas et al… 26

533 Table 3. Mixed-effects model for the effect of treatment, chlorine concentration, temperature,

534 season, exposure time, and their two-way interactions, on cell concentrations. Standard error is

535 given in parentheses.

Predictor β pIntercept 24.70 (0.54) <0.001Concentration -1.30 (0.11) <0.001Temperature -3.48 (0.64) <0.001Season 10.25 (0.53) <0.001Log(Hour + 1) -3.13 (0.21) <0.001Treatment -4.39 (0.50) <0.001Treatment:log(Hour+1) -5.24 (0.18) <0.001Treatment:Temperature 4.26 (0.64) <0.001Log(Hour +1):Concentration -0.60 (0.05) <0.001Log(Hour+1):Temperature 0.72 (0.16) <0.001Concentration:Season 0.27 (0.14) <0.05Temperature:Season -4.53 (0.45) <0.001

536

537

Page 26 of 40



Draft

Casas et al… 27

538 Figures

539 Fig. 1. Measured free chlorine concentrations immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12

540 (t12), 24 (t24), and 48 (t48) hours after chlorine exposure for winter phytoplankton cells maintained

541 at 2 °C and 18 °C. Error bars indicate the Standard Error.

542

543 Fig. 2. Measured free chlorine concentrations immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12

544 (t12), 24 (t24), and 48 (t48) hours after chlorine exposure for summer phytoplankton cells

545 maintained at 2 °C and 22 °C. Error bars indicate the Standard Error.

546

547 Fig. 3. Phytoplankton cells reduction (%) before (tb), immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8

548 (t8), 12 (t12), 24 (t24), and 48 (t48) hours after chlorine exposure for winter phytoplankton cells

549 maintained at 2 °C and 18 °C.

550

551 Fig. 4. Phytoplankton cells reduction (%) before (tb), immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8

552 (t8), 12 (t12), 24 (t24), and 48 (t48) hours after chlorine exposure for summer phytoplankton cells

553 maintained at 2 °C and 22 °C.

554

555 Fig. 5. Ratio Fv/Fm based on variable fluorescence for winter phytoplankton cells maintained at

556 2 °C and 18 °C, immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12 (t12), 24 (t24), and 48 (t48)

557 hours after chlorine exposure for four chlorine concentrations. The dotted line indicates the

558 threshold above which cells are considered photosynthetically efficient. Error bars indicate the

559 Standard Error.

560

Page 27 of 40



Draft

Casas et al… 28

561 Fig. 6 Ratio Fv/Fm based on variable fluorescence for summer phytoplankton cells maintained at

562 2 °C and 22 °C, immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12 (t12), 24 (t24), and 48 (t48)

563 hours after chlorine exposure for four chlorine concentrations. The dotted line indicates the

564 threshold above which cells are considered photosynthetically efficient. Error bars indicate the

565 Standard Error.

566

Page 28 of 40



Draft

Fig. 1. Measured free chlorine concentrations immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12 (t12), 24 (t24), and 48 (t48) hours after chlorine exposure for winter phytoplankton cells maintained at 2 °C and

18 °C. Error bars indicate the Standard Error.

177x175mm (300 x 300 DPI)

Page 29 of 40



Draft

Fig. 2. Measured free chlorine concentrations immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12 (t12), 24 (t24), and 48 (t48) hours after chlorine exposure for summer phytoplankton cells maintained at 2 °C and

22 °C. Error bars indicate the Standard Error

177x175mm (300 x 300 DPI)

Page 30 of 40



Draft

Fig. 3. Phytoplankton cells reduction (%) before (tb), immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12 (t12), 24 (t24), and 48 (t48) hours after chlorine exposure for winter phytoplankton cells maintained at 2 °C

and 18 °C.

177x175mm (300 x 300 DPI)

Page 31 of 40



Draft

Fig. 4. Phytoplankton cells reduction (%) before (tb), immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12 (t12), 24 (t24), and 48 (t48) hours after chlorine exposure for summer phytoplankton cells maintained at 2

°C and 22 °C.

177x175mm (300 x 300 DPI)

Page 32 of 40



Draft

Fig. 5. Ratio Fv/Fm based on variable fluorescence for winter phytoplankton cells maintained at 2 °C and 18 °C, immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12 (t12), 24 (t24), and 48 (t48) hours after chlorine

exposure for four chlorine concentrations. The dotted line indicates the threshold above which cells are considered photosynthetically efficient. Error bars indicate the Standard Error.

166x132mm (300 x 300 DPI)

Page 33 of 40



Draft

Fig. 6 Ratio Fv/Fm based on variable fluorescence for summer phytoplankton cells maintained at 2 °C and 22 °C, immediately after (t0), 1 (t1), 2 (t2), 4 (t4), 8 (t8), 12 (t12), 24 (t24), and 48 (t48) hours after

chlorine exposure for four chlorine concentrations. The dotted line indicates the threshold above which cells are considered photosynthetically efficient. Error bars indicate the Standard Error.

167x133mm (300 x 300 DPI)

Page 34 of 40



Draft

Title: Effect of temperature on chlorine treatment for elimination of freshwater phytoplankton in ballast water:

bench scale test

Oscar Casas-Monroy1*, Julie Vanden Byllaardt2, Johanna Bradie1, Andrea Sneekes3, Klaas Kaag3, and Sarah A. Bailey1

Table A1. Mean phytoplankton abundance (cells mL-1) and their corresponding, standard Error (SE) and percentage mortality (%), for

winter and summer, for each treatment, from tb = time before treatment to 48 hours after chlorine exposure.

Treatment Winter Summer

Temperature ( C) Time points Controls

Mean abundance (cells mL-1) SE

Phytoplankton mortality (%) Controls Mean abundance

(cells mL-1) SEPhytoplankton mortality (%)

0.02 ppm2 Tb 1262 1262 47.2 0.00 533 533 15.8 0.00

T0 1264 834 33.6 34.00 533 514 12.3 3.57T1 1183 766 65.6 35.23 503 315 37.4 37.36T2 1121 670 91.8 40.20 487 249 25.4 48.93T4 1113 628 98.9 43.62 484 231 22.7 52.29T8 1150 626 85.5 45.56 480 212 22.0 55.92T12 1136 448 66.8 60.58 519 169 15.4 67.50T24 1211 445 40.1 63.26 512 170 17.3 66.70T48 1060 420 18.9 60.43 502 143 14.5 71.53

18/22 Tb 701 701 45.0 0.00 451 451 27.2 0.00

Page 35 of 40



Draft

T0 699 706 50.4 -1.01 451 433 27.6 4.05T1 702 637 42.7 9.29 435 337 18.4 22.41T2 680 637 46.9 6.34 433 300 14.6 30.55T4 659 538 40.3 18.31 431 279 11.3 35.40T8 642 424 27.9 33.87 437 238 16.9 45.49T12 582 359 11.9 38.22 467 230 11.5 50.64T24 587 277 12.3 52.90 443 177 15.1 60.09T48 537 202 13.0 62.35 411 131 12.4 68.13

0.22 ppm2 Tb 1262 1262 47.2 0.00 533 533 15.8 0.00

T0 1264 784 21.4 38.01 533 444 26.4 16.70T1 1183 671 77.3 43.31 503 240 48.2 52.33T2 1121 613 100.2 45.30 487 207 28.0 57.50T4 1113 513 104.1 53.93 484 191 24.7 60.48T8 1150 485 70.9 57.83 480 173 20.4 64.04T12 1136 350 58.8 69.16 519 155 16.7 70.15T24 1211 451 31.0 62.74 512 147 17.4 71.33T48 1060 259 37.0 75.59 502 126 14.2 74.88

18/22 Tb 701 701 45.0 0.00 451 451 27.2 0.00T0 699 663 43.6 5.05 451 395 17.1 12.38T1 702 568 25.9 19.16 435 275 10.5 36.76T2 680 531 40.8 21.93 433 241 7.9 44.38T4 659 508 32.9 22.99 431 226 11.2 47.52T8 642 413 24.0 35.59 437 201 10.0 54.13T12 582 335 29.8 42.33 467 180 11.3 61.37T24 587 277 14.2 52.75 443 148 10.3 66.66T48 537 183 21.4 65.90 411 111 9.3 72.98

0.7 ppm2 Tb 1262 1262 47.2 0.00 533 533 15.8 0.00

Page 36 of 40



Draft

T0 1264 869 50.9 31.23 533 391 29.3 26.70T1 1183 723 57.9 38.85 503 207 30.6 58.80T2 1121 679 84.4 39.42 487 170 24.3 65.16T4 1113 561 92.2 49.58 484 153 18.7 68.41T8 1150 494 57.3 57.01 480 143 18.0 70.16T12 1136 354 46.5 68.85 519 133 18.2 74.46T24 1211 410 21.3 66.15 512 120 16.7 76.53T48 1060 303 63.8 71.42 502 102 15.4 79.66

18/22 Tb 701 701 45.0 0.00 451 451 27.2 0.00T0 699 643 38.4 7.92 451 387 21.7 14.17T1 702 535 32.9 23.83 435 277 12.2 36.30T2 680 499 47.7 26.53 433 240 9.7 44.51T4 659 443 39.0 32.86 431 227 12.6 47.44T8 642 383 21.7 40.32 437 198 14.2 54.74T12 582 290 24.3 50.09 467 188 11.1 59.70T24 587 262 9.2 55.45 443 157 12.6 64.64T48 537 172 20.6 68.05 411 115 9.6 72.03

1.2 ppm2 Tb 1262 1262 47.2 0.00 533 533 15.8 0.00

T0 1264 834 59.4 34.03 533 354 20.1 33.58T1 1183 636 49.8 46.27 503 178 28.7 64.59T2 1121 596 64.4 46.83 487 138 23.1 71.61T4 1113 501 69.8 54.98 484 125 19.1 74.17T8 1150 446 55.0 61.22 480 106 15.3 77.84T12 1136 335 29.7 70.54 519 107 16.6 79.39T24 1211 366 22.5 69.75 512 93 16.6 81.73T48 1060 279 64.2 73.71 502 82 14.1 83.58

18/22 Tb 701 701 45.0 0.00 451 451 27.2 0.00T0 699 615 40.6 11.93 451 354 14.7 21.57T1 702 487 32.3 30.61 435 235 14.9 45.98

Page 37 of 40



Draft

T2 680 421 47.5 38.09 433 201 13.5 53.65T4 659 377 38.1 42.83 431 190 15.8 56.04T8 642 348 25.1 45.81 437 162 16.3 63.00T12 582 255 27.2 56.20 467 149 13.0 68.03T24 587 222 12.3 62.15 443 128 11.9 71.22T48 537 171 11.6 68.12 411 97 11.1 76.37

2.2 ppm2 Tb 1262 1262 47.2 0.0 533 533 15.8 0.00

T0 1264 763 61.8 39.61 533 305 11.2 42.86T1 1183 541 72.2 54.30 503 116 12.7 76.84T2 1121 506 77.7 54.83 487 84 11.5 82.75T4 1113 392 67.9 64.82 484 77 11.2 84.11T8 1150 360 60.1 68.73 480 66 10.1 86.31T12 1136 267 46.6 76.47 519 65 10.9 87.43T24 1211 281 40.6 76.83 512 60 10.9 88.18T48 1060 229 54.8 78.38 502 55 10.8 89.12

18/22 Tb 701 701 45.0 0.00 451 451 27.2 0.00T0 699 583 40.9 16.49 451 319 12.6 29.35T1 702 463 42.3 34.09 435 179 12.3 58.81T2 680 373 43.1 45.05 433 146 12.0 66.30T4 659 349 39.9 47.12 431 135 14.3 68.73T8 642 291 22.7 54.58 437 121 13.6 72.25T12 582 218 33.6 62.51 467 113 12.6 75.70T24 587 214 12.7 63.53 443 99 11.4 77.54T48 537 145 12.6 73.02 411 80 9.8 80.59

3.0 ppm2 Tb 1262 1262 47.2 0.00 533 533 15.8 0.00

T0 1264 692 63.4 45.30 533 289 11.2 45.77T1 1183 480 52.9 59.38 503 97 7.7 80.65

Page 38 of 40



Draft

T2 1121 396 60.3 64.66 487 62 5.6 87.25T4 1113 320 52.7 71.26 484 54 5.4 88.91T8 1150 285 49.8 75.24 480 40 6.3 91.75T12 1136 129 12.0 88.67 519 42 6.3 91.95T24 1211 218 35.2 82.01 512 38 6.1 92.48T48 1060 165 31.8 84.42 502 34 6.7 93.28

18/22 Tb 701 701 45.0 0.00 451 451 27.2 0.00T0 699 486 26.6 30.38 451 313 9.0 30.56T1 702 395 28.7 43.73 435 151 9.9 65.15T2 680 255 23.3 62.43 433 116 9.6 73.29T4 659 229 23.1 64.45 431 103 11.6 76.22T8 642 228 22.6 66.07 437 87 13.8 80.05T12 582 197 20.2 71.85 467 86 10.1 81.61T24 587 134 11.7 77.23 443 76 8.6 82.90T48 537 99 16.9 81.62 411 57 10.3 86.19

5.0 ppm2 Tb 1262 1262 47.2 0.00 533 533 15.8 0.00

T0 1264 524 60.8 58.59 533 211 6.4 60.50T1 1183 341 41.3 71.13 503 46 1.4 90.75T2 1121 217 32.9 80.65 487 24 1.2 95.05T4 1113 217 29.9 80.53 484 20 1.0 95.84T8 1150 192 29.0 83.30 480 12 0.6 97.41T12 1136 122 13.9 89.27 519 6 0.4 98.82T24 1211 169 28.7 86.05 512 3 0.5 99.48T48 1060 127 29.1 87.99 502 0 0.1 99.99

18/22 Tb 701 701 45.0 0.00 451 451 27.2 0.00T0 699 397 48.9 43.16 451 218 12.6 51.73T1 702 328 34.5 53.37 435 81 5.1 81.39T2 680 242 30.9 64.34 433 38 2.9 91.24T4 659 186 19.5 71.85 431 26 4.7 93.88

Page 39 of 40



Draft

T8 642 180 17.9 71.94 437 15 2.1 96.66T12 582 155 25.9 73.27 467 9 2.2 98.10T24 587 125 15.5 78.68 443 4 1.2 99.01T48 537 92 17.5 82.93 411 3 1.3 99.15

Page 40 of 40

