The influence of sampling method and season on modeling of selenium into coldwater fish and implications on tissue-based water quality benchmarks

Selenium (Se) contamination of aquatic ecosystems has led to the local extirpation of some Se-sensitive fish species. Although Se exposure occurs primarily via diet, considerable uncertainty lies in modeling Se transfer and bioaccumulation from sediment, detritus, and/or periphyton through benthic macroinvertebrates (BMI) to fish. Here we estimated Se concentrations in four coldwater fish species (northern pike, white sucker, lake whitefish, and ninespine stickleback) inhabiting boreal lakes downstream from a uranium mill in northern Canada. In addition, we evaluated the potential effects of BMI and periphyton sampling methods (artificial substrates vs. grab samples), seasons (summer vs. winter), and models (USEPA vs. Assessment of the Dispersion and Effects of Parameter Transport) on the estimated Se concentrations in fish tissue. Results were compared with site-specific benchmarks and observed Se concentrations in resident fish. In summer 2019, periphyton and BMI were sampled at 10 sampling stations (two in Vulture Lake and eight in McClean Lake) using artificial substrates (n = 4) and sediment grab samples (n = 3). In winter 2021, samples were collected in McClean Lake (n = 3) through ice holes using a sediment grab sampler. Estimated Se concentrations in fish tissue depended on the surface sediment or periphyton Se concentrations used in the models. At Vulture Lake, Se concentrations in northern pike muscle estimated using the grab sample data (17.3 ± 11.5 µg/g DW), but not the artificial substrates (34.5 ± 1.2 µg/g DW), were comparable with the observed mean concentration (19.0 ± 1.4 µg/g DW) in this species. At McClean Lake, Se body burdens in forage fish estimated using data from both sampling methods were comparable with measured data. Significantly lower mean whole-body Se concentrations were estimated for all fish species in winter (1.0 ± 0.3 µg/g DW) relative to summer (4.8 ± 1.6 µg/g DW). Further investigation is necessary to understand how potential seasonal shifts in dietary Se exposure relate to fish reproduction and early life stages.