Sampling method and season influence selenium dynamics at the base of a boreal lake food chain

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.

Laboratory and In situ Selenium Bioaccumulation Assessment in the Benthic Macroinvertebrates Hyalella azteca and Chironomus dilutus

Selenium (Se) bioaccumulation and toxicity in aquatic vertebrates have been thoroughly investigated. Limited information is available on Se bioaccumulation at the base of aquatic food webs. In this study, we evaluated Se bioaccumulation in two benthic macroinvertebrates (BMI), Hyalella azteca and Chironomus dilutus raised in the laboratory and caged in-situ to a Canadian boreal lake e (i.e., McClean Lake) that receives continuous low-level inputs of Se (< 1 μg/L) from a uranium mill. Additional Se bioaccumulation assays were conducted in the laboratory with these BMI to (i) confirm field results, (ii) compare Se bioaccumulation in lab-read and native H. azteca populations and (iii) identify the major Se exposure pathway (surface water, top 1 cm and top 2-3 cm sediment layers) leading to Se bioaccumulation in H. azteca. Field and laboratory studies indicated overall comparable Se bioaccumulation and trophic transfer factors (TTFs) in co-exposed H. azteca (whole-body Se 0.9-3.1 µg/g d.w; TTFs 0.6-6.3) and C. dilutus (whole-body Se at 0.7-3.2 µg Se/g d.w.; TTFs 0.7-3.4). Native and lab-reared H. azteca populations exposed to sediment and periphyton from McClean Lake exhibited similar Se uptake and bioaccumulation (NLR, p = 0.003; 4.1 ± 0.8 µg Se/g d.w), demonstrating that lab-reared organisms are good surrogates to assess on-site Se bioaccumulation potential. The greater Se concentrations in H. azteca exposed to the top 1-3 cm sediment layer relative to waterborne exposure, corroborates the importance of the sediment-detrital pathway leading to greater Se bioaccumulation potential to higher trophic levels via BMI.

Optimizing backwash control using data on seasonal changes in the invertebrate community of granular activated carbon filters

Problems associated with the colonization and leakage of invertebrates in the granular activated carbon (GAC) filters of waterworks have received increased attention in recent years. To study the effect of environmental factors and water quality on invertebrate abundances, and the backwash control for minimizing invertebrate abundance. A survey of the invertebrate community of GAC filters was carried out monthly from March 2021 to May 2022. A pilot-scale GAC system established in the laboratory alongside a lake, with a volume of 35.3 L. 45 invertebrate species were detected, and 40 of these were rotifers. Significant variation in abundance was observed among seasons before and after GAC filtration, the average invertebrate abundance in the inlet water was 11.1 times that in the filtrate. The GAC filter contained invertebrates that might be responsible for the large number of organisms in the filtrate. Invertebrate abundance in the GAC filter decreased gradually with the carbon layer depth, which the mean invertebrate abundances were 6,926, 5,232, and 3818 ind./kg in the top layer (TL), middle layer (ML), and bottom layer (BL), respectively. Invertebrate abundance was correlated with water temperature and varied seasonally. Among eight water quality parameters, chlorophyll a (Chla) and the total plate count (TPC) were most significantly correlated with invertebrate abundance. According to the statistical modeling and the optimization process of response surface methodology (RSM). The predicted optimal values were a flow rate of 6.36 L/h, a backwash cycle of 3.26 d, and a backwash intensity of 14.97 L/(m2·s) for a minimum invertebrate abundance of 3013 ind./kg in the GAC filter. To maintain invertebrate abundance within an acceptable range, some of these measures might need to be modified depending on the actual conditions.