Acute toxic effects of hydrogen peroxide , used for salmon lice treatment , on the survival of polychaetes Capitella sp . and Ophryotrocha spp

The amount of hydrogen peroxide (H2O2) used in the treatment of salmon lice in Norwegian salmon farming increased from 308 tons in 2009 to 43 246 tons in 2015. For 2016 and 2017, however, the consumption was reduced to 26 597 and 9277 tons, respectively. The use of this compound may have negative impacts on benthic fauna underneath the fish farms and, in particular, on polychaetes, which can be found in large numbers at the bottom under fish farms where they play a key role in the turnover of organic waste from the farm. The tolerance of Capitella sp. and Ophryotrocha spp. to a 1 h exposure to H2O2 (0, 100, 200, 400, 800, 1200 and 1800 mg l−1) was evaluated. The recommended dose for treatment of the salmon is 1800 mg l−1. Following exposures, the polychaetes were reintroduced into clean sea water. Both polychaete species experienced high cumulative mortality during a 72 h post-exposure period. The mortality showed to be dose dependent, with the highest dose giving the highest mortality. The 50% lethal concentration (LC50) of Capitella sp. was significantly higher than the LC50 of Ophryotrocha spp. at the same exposure time (p < 0.05). The 50% lethal time of Capitella sp. was significantly longer than that of Ophryotrocha spp. at the same concentration (p < 0.05). The results show that 1 h exposures to H2O2 at all the tested concentrations had irreversible negative effects on both polychaete species.

H 2 O 2 is added to the cage to a final bath concentration of 1500 to 1800 mg l −1 depending on temperature, and the exposure time is 20 to 30 min.Following treatment, the tarpaulin is removed and the released H 2 O 2 may disperse both vertically and horizontally.Since a bath solution of H 2 O 2 is slightly heavier than the surrounding sea water, modeling has shown that the plume may sink after release if the water column is homogeneous (Refseth et al. 2016).This was confirmed by Fagereng (2016), who, in a field investigation, found up to 724 mg l −1 of H 2 O 2 at a depth of 60 m and detectable concentrations even at 130 m.Coupled with a predicted no-effect concentration of 0.01 mg l −1 determined for H 2 O 2 in water (Institute for Health and Consumer Protection in Finland 2003), the results of Fagereng (2016) suggest that the use of H 2 O 2 in salmon farming may be harmful to nontarget organisms located near salmon farms and, given the right conditions, benthic organisms like polychaetes may also be exposed.
H 2 O 2 has long been regarded as an environmentally friendly salmon lice agent because it readily dissociates to water and oxygen.However, Fagereng (2016) calculated degradation half-lives of 28 and 3.5 d of H 2 O 2 in seawater at temperatures of 8.7 and 12°C, respectively.Furthermore, it has been shown that low concentrations (< 0.170 mg l −1 ) of H 2 O 2 affect the physiology of aquatic animals, such as antioxidant enzyme activities in the polychaetes Arenicola marina and Nereis (Hediste) diversicolor (Abele-Oeschger et al. 1994, Buchner et al. 1996).Effects on oxygen consumption, catalase and glutathione peroxidase activity were further seen in the polychaete Laeonereis acuta when exposed to concentrations of 0.34 and 1.7 mg l −1 for up to 10 d (da Rosa et al. 2008).The acute effects of high concentrations of H 2 O 2 on aquatic invertebrates are largely unknown, but Fagereng (2016) showed that 1 h exposure of pink shrimps Pandalus montagui to 170 mg l −1 led to reduced flight response even after a 24 h recovery period.A 1 h exposure of copepods Calanus sp.gave a 50% lethal con centration (LC 50 ) of less than 5% of the recommended dose of 1700 mg l −1 (Escobar-Lux 2016).On the other hand, a 1 h exposure gave an LC 50 higher than 1700 mg l −1 for sand shrimp Crangon septemspinosa, rock pool shrimp Palaemon elegans, chameleon shrimp Praunus flexuosus and adult American lobster Homarus americanus (Burridge et al. 2014, Brokke 2015).Hence, major differences in sensitivity between species are seen.Therefore, there is a need for more knowledge about the effect of H 2 O 2 exposure on non-target organisms, especially for benthic species.
Polychaetes are naturally abundant in benthic habitats under fish farms and in other types of anthropogenically modified estuaries (Kutti et al. 2007, Dafforn et al. 2013, Bannister et al. 2014).Opportunistic polychaetes that are adapted to nutrient-rich habitats and commonly found underneath fish farms located over hard bottom in Norway include Vigtorniella ardabilia and Ophryotrocha spp.(Paxton & Davey 2010, R. Bannister pers. comm.) and over soft-sediment areas, Capitella sp.(Kutti et al. 2007, Dean 2008).The polychaetes are important in environmental recovery by consuming and transforming the organic materials deposited from the fish farm (Dean 2008).Because these species live near the fish farms, they may be exposed to agents originating from activities at the farm, including salmon lice treatment.The objective of this study was therefore to find the limit of tolerance of Capitella sp. and Ophryotrocha spp. to short time exposure to H 2 O 2 .This will contribute to the evaluation of the effects of sea lice drugs on the natural environment surrounding fish cages.

Animal collection and acclimatization
Capitella sp. were collected by grab sampling (250 cm 2 ) underneath a fish farm located at Austevoll, Norway.Ophryotrocha spp.were collected at a fish farm in Hardangerfjord, Norway, using artificial plastic grass mounted in an iron frame of 1.2 × 1.2 × 0.1 m and deployed underneath a fish cage for 2 wk.The collected polychaetes under the fish farms were representative species.At both fish farms, alternative methods (fresh water, increased water temperature) had recently been used for delousing purposes.Directly after sampling, polychaetes were placed in boxes containing sea water collected from about 150 m depth.The boxes, supplied with air, were transported to the laboratory at Austevoll Research Station (Institute of Marine Research, Norway).Capitella sp.specimens were placed in four 100 l tanks, with 1 kg of glass beads (6 mm diameter) in each tank mimicking artificial benthic substrate.The Ophryo trocha spp.were placed in 100 l tanks, each supplied with 5 stones of about 300 g serving as substrate.The stones facilitated aggregation of Ophryo trocha spp.and provided a rough substrate to attach mucus strings, mimicking a hard-bottom substrate.Tanks were supplied with a seawater flow of 1150 to 1500 ml min −1 from 150 m depth holding a temperature of 8 to 9°C.The polychaetes were acclimatized for 5 d and fed ground salmon pellets produced by Skretting once a day.The tanks were kept in darkness during the acclimation period, except during feeding.

Experimental design
Polychaetes were exposed to 6 nominal concentrations of H 2 O 2 (100, 200, 400, 800, 1200, 1800 mg l −1 ) for 1 h, where the highest concentration is equal to the recommended dose used for treatment.Concentrations were prepared by diluting the stock formulation (Nemona 49, 5%, Akzo Nobel) with sea water to the desired concentration for each treatment.The polychaetes (> 50 individuals, estimated from pre-calculated volume per numbers) were transferred to 2 l beakers containing the decided concentration of H 2 O 2 .Beakers without H 2 O 2 served as controls.Three replicate groups were used for each concentration, including control groups.Following exposure (1 h), the H 2 O 2 solution in the beaker was replaced with clean water and a continuous flow (150−180 ml min −1 ) of sea water established.The number of dead animals was recorded at 1, 6, 12, 24, 48 and 72 h from the start of H 2 O 2 exposure; the number of remaining survivors was also counted at 72 h.Beakers were kept in the dark during the experimental period.

Statistical analysis
The 50% lethal time (LT 50 ) is the time where 50% of the organisms have died after exposure to a toxic substance or stressful condition.The LC 50 refers to the concentration for half the population to die from a treatment or exposure.The LC 50 and LT 50 values were calculated by the Bliss Probit Method (Sprague 1969).Data were analyzed using the SPSS for Windows (Version 13.0) statistical package.The calculated LC 50 and LT 50 values are presented as mean ± SD, unless stated otherwise.The differences in mortalities for each treatment and time intervals were assessed with 1-way ANOVA followed by Duncan's multiple range tests for post hoc pairwise comparisons.A dependent t-test was applied to detect any differences between LT 50 and LC 50 for the 2 species.
Curve estimation was used to analyze the relationship between LC 50 and time and between LT 50 and concentrations.Differences were statistically significant if p < 0.05.

Mortality
For Capitella sp., no mortality was seen in the control groups.In the H 2 O 2 -exposed groups, the acute mortality after 1 h exposure was dose dependent, with the 2 highest doses giving a mortality of > 60%.
A delay in mortality, on the other hand, was seen for the 100, 200 and 400 mg l −1 doses (Fig. 1a).The cumulative mortality increased gradually throughout the experimental period of 72 h, reaching over 90% for all doses except for 100 mg l −1 , which reached 76% (Fig. 1a).
The sensitivity of Ophryotrocha spp. to H 2 O 2 was significantly higher than that of Capitella sp.A 1 h exposure resulted in acute mortality for all doses, reaching 100% for the 1200 and 1800 mg l −1 doses and 20% for the 100 mg l −1 dose (Fig. 1b).After 72 h, the cumulative mortalities were nearly 100% for all doses (Fig. 1b).Some mortality was registered in the control beaker but was less than 10% after 72 h.

LC 50 and LT 50
A significant relation was found between LC 50 and exposure time for both Capitella sp. and Ophryotrocha spp.(p < 0.05, Fig. 2).The curve estimation analysis for the LC 50 of Capitella sp.showed a higher Similarly, a significant relation was observed between LT 50 and H 2 O 2 concentrations for both species (p < 0.05, Fig. 3).The LT 50 for Capitella sp. with 76, 32, 33 and 11 h for the doses 100, 200, 400 and 800 mg l −1 , respectively, was significantly longer compared to Ophryotrocha spp., with an LT 50 of 24 and 4 h for doses of 100 and 200 mg l −1 , respectively (t-test, p < 0.05).The results thus indicate that more than 50% of the Ophryotrocha spp.population would not survive 1 h exposure to H 2 O 2 if the concentration exceeded 400 mg l −1 .For Capitella sp., a concentration of 1200 mg l −1 would result in 50% mortality within 1 h.

DISCUSSION
Even with an exposure time of only 1 h, both Capitella sp. and Ophryotrocha spp.showed low acute tolerance to the recommended dose of H 2 O 2 used for delousing.Both species also uncovered limited capacity to recover after exposure to all concentrations tested.An observed effect in both species was change in skin color.The skin of Capitella sp.turned from red to gray during the exposure, and the skin of Ophryotrocha spp.turned from light red to white.This was particularly clear for the high concentrations.Most of the polychaetes that were alive after exposure did not survive the recovery period.Therefore, it seems that the damage from H 2 O 2 exposure is irreversible in both species and leads to high mortality even at doses that are realistic and ecologically relevant.Further studies should therefore include even lower doses of H 2 O 2 and be combined with a longer recovery period and studies of sublethal effect parameters.The mortality after 1 h exposure was considerably lower in Capitella sp.than in Ophryotrocha spp.However, this difference was reduced at the end of the experiment, as both species experienced a substantial mortality in the recovery period.This highlights the importance of including an extended recovery period when studying compounds like H 2 O 2 .Low H 2 O 2 concentrations (< 2.0 mg l −1 ) have previously been shown to cause adverse effects on polychaetes (Abele-Oeschger et al. 1994, Buchner et al. 1996, da Rosa et al. 2008).As Capitella sp.inhabit benthic sediments and may show behavioral avoidance via burrowing, this may affect the exposure to H 2 O 2 discharged from the farms.However, analytical challenges make it difficult to definitively determine whether dis charged H 2 O 2 might infiltrate the sediment, making burrowing less advantageous.More work should therefore be done to study the effects of H 2 O 2 on the Capitella sp. when settled in a sediment.For the Ophryotrocha spp.living on the surface of  2017) used Perhydrol, a 30% pro-analysis product from Merck, in their study.Whether there may be differences in the toxicity, first between the 2 antisea lice products and second between those products and the pro-analysis product, has to our knowledge not been investigated.
Following treatment, the H 2 O 2 is released and will disperse horizontally and vertically, if the water column is homogeneous, since a bath solution of H 2 O 2 is slightly heavier than the surrounding sea water (Refseth et al. 2016). Fagereng (2016), in one of a very limited number of field studies, reported vertical distribution of H 2 O 2 , finding concentrations of 271 to 723 mg l −1 at a depth of 60 m for nearly 20 min at one sampling station but also horizontal distribution, where the drug was found in the upper 30 m and at concentrations up to 69 mg l −1 .The discharged H 2 O 2 from fish farms is therefore likely to be harmful for the polychaetes underneath and in the proximity of the fish farms.
Referring to Sprague (1971), the safe concentration of H 2 O 2 is assumed to represent 1% of LC 50 at 72 h for estimation of chronic, sublethal and cumulative H 2 O 2 toxicity.Using this assumption, the safe concentrations of H 2 O 2 to Capitella sp. and Ophryotrocha spp.will be 1.59 and 0.64 mg l −1 , respectively.H 2 O 2 has long been regarded as the most environmentally friendly anti-salmon lice agent.This study demonstrates nonetheless that the potential risk of H 2 O 2 to aquatic organisms is obvious and serious and therefore requires more attention in research and legislation than previously assumed.It is thereby important to define relevant concentrations of H 2 O 2 in the environment.Further research should also include non-lethal stress responses and different life stages of the organisms tested.

Fig. 2 .
Fig. 2. Lethal concentration (LC 50 , mg l −1 ) of the polychaete species Capitella sp. and Ophryotrocha spp.exposed to hydrogen peroxide at increasing time intervals after exposure.The equations stem from the results of curve estimation analysis of the relationship between LC 50 and time