2.1 Chemicals
Three deicing agents were used in the current study. Laboratory grade sodium chloride (CAS: 7647-14-5) was purchased from Fisher Scientific (Pittsburg, PA, USA). Snow Joe Melt Enviro-blend Ice Melter w/CMA, (Snow Joe, LLC, Edison NJ, USA) and Morton Safe-T-Salt (Blue) (Morton International, Inc., Chicago, IL USA), were selected based on their availability and popularity of local use. For simplicity, Snow Joe Melt Enviro-blend Ice Melter w/CMA and Morton Safe-T-Salt (Blue) will be referred to as RS1 and RS2 respectively. According to the product label RS1contained 85-99.9% sodium chloride with 0.01-5% magnesium chloride and contained CMA (0.01-5%) and was dyed green. RS2 listed yellow prussiate of soda as an additional ingredient and was dyed blue. Both salts were described as “pet safe”. The four other salts used for comparison were Industrial Strength Ice Melt (Scotwood Industries, Overland Park, KS, USA), Frosty’s Nightmare (Dart Seasonal Products, Spotswood, NJ, USA), Core Multi-Power Ice Melt (Midwest Salt, LLC, West Chicago, IL, USA) and Morton Fast Acting Ice Melt (Morton International, Inc., Chicago, IL USA). These also varied in color (mostly blue, green or white) but contained no other active ingredients besides salts. All other reagents (e.g., salts used to prepare water) were purchased from Fisher Scientific (Pittsburg, PA, USA).
2.2 Toxicity bioassays
Standard 24 and 96 h water-only bioassays were conducted using three juvenile aquatic macroinvertebrates: Hyalella azteca (7-10 d old), Hexagenia rigida (approximately 8 weeks old or 5 mg) and Chironomus dilutus (early 3rd instar). Each species was tested for both exposure durations for each of the three salts described above. All organisms were obtained from existing laboratory cultures at Alma College (Alma, MI, USA). Hyalella azteca and Chironomus dilutus were cultured using methods adapted from the U.S. Environmental Protection Agency (U.S. EPA 2000). Hexagenia rigida were cultured from eggs collected from Lake Erie using previously established methods (Harwood et al. 2014).
The dosing concentrations used were determined through preliminary experiments with a wider range of concentrations using methods similar to those described below. Five dosing concentrations and a negative control of moderately hard water (MHW) (Smith et al. 1997) were used in each test. Due to the large size of some of the road salt crystals, the salts were reduced in size using a ceramic mortar and pestle to increase the rate at which they dissolved. Salts were weighed on a Mettler Toledo (Columbus, OH, USA) balance (± 0.1 mg). The salt was added to MHW and mixed on a stir plate set at medium-high speed until the salts were completely dissolved. Four replicate 600 ml beakers with 500 ml of salt solution or control MHW were used for each of the six concentrations (Table 1). Each species and exposure duration were tested separately with different salt concentrations. Ten organisms were added to each replicate. To beakers containing H. azteca a 2.5 cm. x 2.5 cm square of stainless-steel screen was added to serve as a substrate for the organisms. Glass artificial burrows constructed from 10 disposable glass pipette tops adhered to a glass microscope slide with aquarium grade silicone were used in Hexagenia exposures as a substrate. Beakers containing C. dilutus contained 25 mL of Fisher Washed Sea Sand (Thermo Fisher Scientific, Waltham, MA, USA) to cover the bottom of the beaker and serve as substrate. Water quality parameters (dissolved oxygen, temperature, conductivity, and pH) were checked pre and post exposure using an Oakton DO 6+ Meter and an Oakton PC 450 Meter (Cole-Palmer, Vernon Hills, IL, USA). All experiments were conducted in Precision 818 Environmental Chambers (Thermo Fisher Scientific, Waltham, MA, USA) maintained at 23±1 °C with a 16 light: 8 dark photoperiod. At the completion of either 24 or 96 h, the surviving animals were enumerated.
2.3 Statistical analysis
The LC50, for each species, duration, and salt was calculated using a Probit analysis in RStudio, Version 4.1.2 (Boston, MA) (RStudio Team, 2021). Chloride ion concentrations among salts were compared using a one-way ANOVA with a Tukey’s HSD post hoc test in RStudio. Nominal salt concentrations (g/L) were used to calculate median lethal levels. Statistical comparisons among LC50 values were done using overlapping confidence intervals (Cumming et al. 2007).
2.4 Comparison of chloride ion concentrations
Four replicates of five solutions (2, 4, 6, 8, and 10 g/L) were prepared using each of the road salts and NaCl. These concentrations were chosen to be comparable to those used in the bioassays. The solutions were made in 30 mL of deionized water in conical vials and homogenized by hand. The Vernier GoDirect Chloride Ion-Sensitive Electrode (Vernier, Beaverton, OR, USA) was used to measure the concentration of chloride ion in each solution. The probe was calibrated using the NaCl solution. The calibration value was used to calculate the percent of chloride ion mass in each solution. The average and standard deviation of the percent chloride ion was determined for each salt at each concentration to determine the consistency of these estimates.