Examination of mercury contamination in water striders and minows inhabiting watersheds in coastal Maine
January 2004 - December 2007
- Maine Outdoor Heritage Fund
- Cooperative Research Units
The primary purpose of this study was to use stream-dwelling biota (e.g., northern two-lined salamander larvae, slimey sculpin) to quickly assess stream mercury contamination. During July 2006 we surveyed 25 streams for larval northern two-lined salamanders and slimey sculpin in Penobscot, Washington, and Piscataquis Counties. We found very few of either species in any of the streams. Given that our purpose was to use easily collected, stream-dwelling biota as a rapid-assessment tool for mercury contamination, we selected alternate, easily captured target species [water striders (Aquarius remigis) and “minnows” (blacknose dace Rhinichthys atratulus; northern redbelly dace Phoxinus eos; pearl dace, Semotilus margarita; creek chub, Semotilus atromaculatus; lake chub, Couesius plumbeus; common shiner, Luxilus cornutus; golden shiner, Notemigonus crysoleucas] that were abundant in these streams. The collected fish (“minnows” hereafter) represent a typical assemblage found in low order streams. We selected water striders because there was precedence for use of this species as a possible mercury sentinel in low order streams. Total mercury concentrations in water striders collected from small streams draining gold mine tailings in New Brunswick, Canada, were correlated with mercury concentrations in brook trout (Salvelinus fontinalis) collected from the same site (Jardine et al. 2005). Ratios of methyl to total mercury were similar in the collected striders and minnows, and striders contained lower mean concentrations of mercury than minnows. Except for two streams (Burdin and Buzzell), methyl and total mercury concentrations were generally similar among the taxa. Methyl and total mercury in the fish assemblage in these two streams exceeded concentrations in striders. Total and methyl mercury concentrations generally were more variable in the fish than striders, which may reflect their longer lifespan (striders <2 years, fish species 2-4 years). There was some inconsistency in relationships of mercury concentrations among streams; samples of both taxa with the greatest methyl and total mercury concentrations were not collected in the same stream. These organisms should not be used, therefore, as a predictor of mercury concentration in the other taxa in a particular stream, but can be a relative indicator of mercury contamination in the stream. Jardine et al. (2005) found a significant relationship between total mercury concentrations in striders and brook trout in New Brunswick streams. This relationship is influenced by the range of concentrations in their study; if samples from the collection site closest to the mine tailings were removed from their dataset, the relationship between trout and strider total mercury concentrations would have been weaker and thus more similar to our results. Although we did not document consistent relationships between strider and minnow mercury concentrations in these streams, the concentrations of these contaminants in the streams we sampled are at least in the range of those reported in New Brunswick, Canada, streams, by Jardine et al. (2005). At least one fish from four of the streams we sampled contained concentrations of total mercury exceeding those reported in brook trout by Jardine et al. (2005), and at least one fish collected in our study from Lowell Brook contained total mercury concentrations that exceeded those in trout collected below the New Brunswick, Canada, mine tailings site. Similarly, at least one of our striders from Lowell Brook contained total mercury concentrations exceeding those measured in striders collected near the mine tailings site. Given our small sample size and geographically and temporally restricted study, it would be prudent to sample striders in more streams in the area and over a longer period. The concentrations of total and methyl mercury in striders and minnows collected from the sampled streams suggest that this contaminant is likely widespread in the foodweb of small streams throughout this region of Maine; population level effects of this level of contamination on biota in these streams remain unknown.