Factors affecting mercury concentrations in Iowa fishes

Mercury contamination in aquatic ecosystems is a global concern due to the health risks of consuming contaminated aquatic organisms, particularly fishes. Mercury concentrations in fishes are highly variable and influenced by a range of biotic and abiotic variables. Currently, factors influencing mercury accumulation in Iowa fishes are not well understood. The Iowa Department of Natural Resources (IDNR) has issued fish consumption advisories for various lakes and river reaches throughout the state. However, relatively few systems, species, and individuals are sampled each year and little is known regarding factors affecting mercury concentrations in Iowa fishes. An understanding of factors regulating mercury concentrations in Iowa fishes would improve mercury monitoring programs and consumption guidelines. The objectives of this study were to (I) evaluate seasonal variation in mercury concentrations in largemouth bass (Micropterus salmoides) from two reservoirs to assess the need for temporally standardized mercury sampling and (II) evaluate the influence of a suite of biotic and abiotic factors on fish mercury concentrations in both river and lake systems. Largemouth bass were intensively sampled from Red Haw and Twelve Mile lakes to evaluate temporal variation in largemouth bass mercury concentrations. Bluegill (Lepomis macrochirus, n = 275), white and black crappie (Pomoxis annularis, n = 112; P. nigromaculatus, n = 203), largemouth bass (n = 502), walleye (Sander vitreus, n = 248), muskellunge (Esox masquinongy, n = 30), and northern pike (E. lucius, n = 45) were collected between April and October, 2013-2015, from natural lakes (n = 6), shallow natural lakes (n = 2), constructed lakes (n = 18), and reservoirs (n = 4) throughout Iowa. Additionally, channel catfish (Ictalurus punctatus, n = 205), flathead catfish (Pylodictis olivaris, n = 123), northern pike (Esox lucius, n = 60), smallmouth bass (Micropterus dolomieu, n = 176), and walleye (Sander vitreus, n = 176) were collected between March and October, 2014-2015, from ten Iowa interior rivers and tested for mercury contamination. Fish were collected from an upstream and a downstream location on six of the rivers to test for intra-river differences in fish mercury concentrations. Various land use, water chemistry, and fish characteristics were gathered and used to explain differences in mercury concentrations within and across lake and river systems. Largemouth bass mercury concentrations varied across months in Red Haw Lake, with the highest concentrations observed during July, and the lowest concentrations observed during October. In contrast, largemouth bass mercury concentrations were similar across months in Twelve Mile Lake. Fish mercury concentrations in Iowa lakes are generally low, with mercury concentrations \u3c0.30 mg/kg for ~90% of fishes collected and mercury concentrations below detectable levels (\u3c0.05 mg/kg) for ~40% of fishes. Multiple linear regression models, sorted by AICc, were used to evaluate factors related to fish mercury concentrations in lakes and rivers. Detected mercury concentrations were highest in muskellunge, northern pike, walleye and largemouth bass, lowest in black and white crappie and bluegill, and positively related to fish length and age. Although not significantly different across all species, females generally had higher mercury concentrations than males. Additionally, pH, lake mean depth, watershed area to lake area ratio, and percent of watershed as forested land, grasslands and open water were positively related to fish mercury concentrations, whereas lake area and percent of watershed as agriculture and developed land were negatively related to mercury concentrations. Finally, detected mercury concentrations were on average 28% higher in shallow natural lakes compared to other lake types. Combined, these factors explained 74% of the variation in detectable fish mercury concentrations. Fish mercury concentrations in Iowa rivers were also generally low (mean = 0.17 mg/kg, N = 740). Mercury concentrations were highest in flathead catfish, northern pike, smallmouth bass, and walleye but lowest in channel catfish. Fish mercury concentrations were positively related to length, age, trophic position and δ13C signatures. Human Threat Index and percent of watershed as open water were negatively related to fish mercury concentrations, whereas percent of watershed as forested land was positively related to fish mercury concentrations. Additionally, phosphorous (mg/L), nitrogen-ammonia (mg/L), and sulfate (mg/L) were weakly negatively related to mercury concentrations, whereas water hardness (as CaCO3, mg/L) was weakly positively related to fish mercury concentrations. Additionally, fishes collected from the Paleozoic Plateau ecoregion had the highest mercury concentrations compared to those collected from other ecoregions across Iowa. Together, these factors explained 70% of the variation in river fish mercury concentrations. Results of this study suggest less impacted watersheds, particularly watersheds with less agricultural impacts, tend to have higher fish mercury concentrations compared to watersheds that have a high proportion of agriculture. This study provides a comprehensive analysis of abiotic and biotic factors influencing fish mercury concentrations in Iowa and may have implications for refining consumption advisories