Silver

Long-term toxicity to fish

Description of key information

Three chronic NOECs were taken forward for use in the derivation of the freshwater PNEC by statistical extrapolation. These were an EC10 of 0.17 µg/L Ag for Oncorhynchus mykiss (Davies et al., 1998), an EC10 of 0.19 µg/L Ag for Salmo trutta (Davies et al. 1998) and an EC10 of 0.39 µg/L Ag for Pimephales promelas (Naddy et al. 2007). A NOEC of 130 µg/L Ag for Menidia beryllina (Ward et al. 2006) is taken forward for the derivation of the marine PNEC.

Additional information

Key data for three fish species are available: rainbow trout (Oncorhynchus mykiss), brown trout (Salmo trutta) and fathead minnow (Pimephales promelas). Key data for O. mykiss and S. trutta are taken from Davies et al. (1998). Davies et al. (1998) undertook six long-term exposures with two species of salmonid to investigate the effects of water quality characteristics on silver toxicity. Salmo trutta (brown trout) and Oncorhynchus mykiss (rainbow trout) were exposed to silver in flow-through tanks under three water “hardness and chloride” regimes (soft, medium and hard). Survival and development were monitored from hatch to 217 days (31 weeks) post-hatch. The test medium for the three “hardness and chloride” regimes was created by diluting well water with reverse osmosis treated well water. As a consequence of this, other physicochemical parameters also varied between the hardness regimes, e.g. DOC and SO₄. EC10 values for chronic S. trutta survival were calculated from the data in the report as 0.19, 0.75 and 1.23 μg/L dissolved silver for soft, medium and hard water, respectively. Similarly, EC10 values for O. mykiss survival were calculated as 0.17, 0.30 and 0.63 μg/L dissolved silver, respectively. Only the results of the soft “hardness and chloride” regime experiments are suitable for use in the calculation of the PNEC as the study indicates a reduction in silver toxicity (of between three- and six-fold) with increasing water “hardness and chloride”, potentially through modified bioavailability of silver.

Additional reliable supporting data for O. mykiss are also available in Davies et al. (1977), Morgan et al. (2005), Dethloff et al. (2007) and Nebeker et al. (1983). However, with the exception of Davies et al. (1977), these studies are not of comparable duration or sensitivity to Davies et al. (1998) and will therefore not be used in the calculation of the PNEC. The data in Davies et al. (1977) are of similar sensitivity to those presented in Davies et al. (1998) but will not used in the calculation of the PNEC as the results are based on total rather than dissolved concentrations of silver.

Key long-term data for fathead minnow (Pimephales promelas) are taken from Naddy et al. (2007). Naddy et al. (2007) investigated the chronic toxicity of silver to fathead minnows at low and enhanced chloride concentrations. Fathead minnows were exposed to silver in soft reservoir water in standard ASTM early life stage (ELS) toxicity tests (up to 34 days) at one of two flow regimes: normal (15 mL test solution/min) and high (50 mL test solution/min). Chloride concentrations were increased in the enhanced chloride treatment by addition of NaCl to increase the concentration from an original 1.9 mg/L to 58.2 mg/L. The authors report MATC values of 0.53, 0.36 and 0.83 µg/L dissolved silver for normal flow/low chloride, high flow/low chloride, and high flow/enhanced chloride treatments, respectively. A similar pattern of relative toxicity was observed in EC10 values for growth calculated from the data in the paper of 0.39, 0.59 and 1.41 μg/L dissolved silver for normal flow/low chloride, high flow/low chloride and high flow/enhanced chloride treatments, respectively. Only the results of the normal flow/low chloride treatment are used in the SSD as the results suggest a combination of high flow and chloride has the potential to reduce chronic toxicity. Additional reliable supporting data are available in Holcombe et al. (1983) but are not used in the SSD as they are based on total silver concentrations.

In the marine environment Ward et al. (2006) conducted tests with the fish Menidia beryllina. The 28-day survival NOECs were 26, 49 and 130 µg/L dissolved silver at 10, 20 and 30‰ salinity, respectively. Klein-MacPhee et al. (1984) report a NOEC of 54 µg/L dissolved silver for growth in winter flounder (Pleuronectes americanus) after an 18 day exposure of embryos. Marine fish appear to be somewhat less sensitive to the effects of silver than freshwater species. This is likely to be due to the speciation of silver in marine / estuarine systems.