Fresh Water Aquatic Life Working Group, Equilibrium Environmental Inc.
December 2014
Executive Summary
The Canadian Council of Ministers of the Environment (CCME) has recently revised the Water Quality Guideline for chloride (from 230 to 120 mg/L for chronic exposure), applicable to freshwater aquatic life receptors where COSEWIC (Committee on the Status of Endangered Wildlife in Canada) Endangered or Special Concern Pelecypoda (clams, mussels) are not present (CCME 2011). In their analysis, toxicity testing data demonstrates that Pelecypoda at the glochidia lifestage can be particularly sensitive to the adverse effects associated with chloride exposure, which was a partial determinant of the guideline revision.
The CCME also identified water hardness as an important factor that modifies the toxicity of chloride towards aquatic life. This phenomenon has been demonstrated in the literature over a range of species, resulting in various other regulatory agencies adopting hardness adjusted guidelines for chloride and other major ions. The state of Iowa has developed acute (287.8 *(hardness)0.205797(sulphate)-0.07452) and chronic (177.87 * (hardness)0.205797(sulphate)-0.07452) algorithms for the adjustment of a chloride water quality guidelines based on hardness (IDNR 2009a). Within Canada, British Columbia has recently established a hardness adjusted guideline for the sulphate ion varying from 128, 218, 309, and 429 mg SO4 /L for very soft (0-30 mg/L), soft / moderately soft (31-75 mg/L), moderately soft / hard (76-180 mg/L), and very hard (181-250 mg/L) water respectively (BCME 2013). Water hardness has also been identified as an ameliorating factor in aquatic guideline policy such as the Australian and New Zealand government’s National Water Quality Management Strategy (ANZECC, 2000). Finally, the effects of hardness amelioration have been incorporated into site specific guidelines for chloride in Canada. Examples include the EKATI diamond mine (guideline calculated as: 124 * ln (hardness) -128) (Rescan 2007) and the Snap Lake diamond mine (De Beers 2013). Given the growing application of hardness derived guidelines to major ion toxicity, an important component of any future research on the toxicity of chloride towards sensitive aquatic organisms is to incorporate varying hardness levels and produce information that can be used to improve the accuracy of guidelines developed on a provincial or national scale.
Although recognized by CCME, as a result of limitations to the long-term dataset, a hardness derived guideline was not established in 2011. Long-term hardness-toxicity data was either not available or did not meet the minimum data requirements for hardness adjusted water quality guidelines at the time that CCME developed their limit in 2011.
The objectives of the present study were to:
An aquatic toxicological database for chloride was established using data from 189 studies, of which 69 represented additional studies not included in the previous CCME (2011) dataset. The dataset developed herein was comprehensive and captured various data, including: effect concentrations; multiple toxicity endpoints; exposure durations; information on other ions in solution (water hardness, major ion concentrations); organism data (taxonomic, life history, geographic distribution); and, environmental conditions (temperature, pH, dissolved oxygen, and light exposure).
Sufficient information was available to construct short-term SSDs to compare the relative toxicities of the four common chloride salts (NaCl, KCl, CaCl2, and MgCl2). Results agreed with the established literature and showed that the relative toxicity of the salt pairings was: KCl > MgCl2 > NaCl ≈ CaCl2, where the toxicity of potassium and magnesium chloride can be attributed to the associated cation rather than chloride. Based on these findings, the remainder of the guideline derivation was based on NaCl and CaCl2 as was conducted by the CCME (2011). Unlike the CCME, CaCl2 data was not pooled with NaCl data, and the protectiveness of the derived guidelines to CaCl2 was assessed separately, which is anticipated to be a conservative procedure in terms of deriving a limit for NaCl as CaCl2 appeared to be slightly less toxic.
For comparative purposes, the initial assessment of the short- and long-term NaCl data involved creating SSDs following the general methods used by CCME (2011), without any adjustment for water hardness. The guideline derived from the unadjusted short-term SSD was 659 mg/L, close to the value of 640 mg/L established by the CCME (2011). This minor disparity might be explained by the different modeling approach (log-Normal model used by the CCME), the five additional species (Hydra attenuate, sp., Pycnopsyche sp., Hyla chrysoscelis, Chironomus xanthus) used in the current data set, or from the higher toxic effect concentration used for the fatmucket clam (L. siliquoidea). The collated effect concentration used for this species was increased from 709 mg Cl- /L (CCME 2011) to 892 mg Cl- /L, with the addition of data from Cope et al. (2008) that was not included by CCME. Because of its relatively low position in the SSD, this shift in effect concentration may have resulted in a higher modeled benchmark concentration. Other changes included limits for C. dubia (1080 to 1151 mg Cl- /L), and D. pulex (1248 to 1597 mg Cl- /L). Overall the 20 mg/L increase was not seen as a significant departure from the results obtained by the CCME (2011).
The guideline derived from the unadjusted long-term SSD was 112 mg/L, 8 mg/L lower than the 120 mg/L value established by the CCME (2011). In both derivations (CCME and herein), the Log-Logistic model was found to best fit the data. The more sensitive guideline calculated herein is the result of new additions to the SSD dataset, along with revised effect concentrations for a number of species located relatively low on the curve. Three data points in particular are important, and likely the primary factors driving the greater sensitivity of the current model. One is a 96h EC10 endpoint of 96 mg Cl- /L for the green algae P. subcapitata (Simmons 2012). This data point represents a departure from the CCME derivation where algae species were only represented high on the curve with effect concentrations > 6,000 mg Cl- /L (See Section 4.1.1 for further discussion). Two data points representing C. dubia and L. minor have also been reduced from 454 to 337 mg Cl- /L and 1171 to 496 mg Cl- /L, respectively, with the addition of data from Elphick et al. (2011a), Lasier and Harden (2010), and Simmons (2012).
In order for short- and long-term hardness adjusted guidelines to be established, chloride toxicity values from different studies must be compared by converting them to a standardized hardness value. Here, the standard hardness of 50 mg/L (CaCO3 equivalents) was used, as has been done in previous guideline derivations for cadmium (CCME 2014, US EPA 2001). Using statistical analysis, empirical relationships were established for both short- and long-term data in order to convert toxicity data to the standard hardness value of 50 mg CaCO3/L. This involved establishing hardness-toxicity slopes for organisms where effect concentrations were available over a wide range of water hardness concentrations. Individual hardness toxicity slopes were then statistically pooled to determine an overall estimation of the hardness-toxicity relationship within the short- and long-term datasets. For a number of organisms, such as the bivalve Lampsilis fasciola, inverse hardness-toxicity relationships were established but could not be included in pooled slope calculations since hardness-toxicity relationships were not investigated over a wide enough range of hardness (sensu Stephan et al. 1985).
Short- and long-term hardness-adjusted (50 mg CaCO3/L) datasets were then assessed using an SSD approach and the best fit cumulative distribution function was utilized for guideline derivation. The derived short-term benchmark concentration and long-term WQG are presented as exponential functions allowing the determination of guidelines based on site specific water hardness concentrations.
The chloride short-term benchmark concentration and long-term WQG for the protection of aquatic life were based on CCME protocols for Type A (statistical – SSD derivation) data (CCME 2007) and utilized endpoints for NaCl, which was the only chloride salt found to satisfy the requirements for Type A analysis. The short-term benchmark was based on the distribution of data from 46 species modeled using a log-logistic function. The long-term WQG was based on the distribution of data from 32 species also modeled using a log-logistic function. These guidelines are presented below, along with the results chloride concentrations at various water hardness concentrations.
Summary of chloride guidelines for the protection of aquatic life
| Short-term exposure (mg/L) | Long-term exposure (mg/L) | |
| Hardness equation | Benchmark = 10[ 0.294 (log(hardness)) + 2.179] | WQG = 10[ 0.473 (log(hardness)) + 1.227] |
Notes: 1) hardness measured as mg/L as CaCO3; 2) the short-term hardness equation is applicable from 28 to 796 mg CaCO3 /L and
should not be applied outside of this range; 3) the long-term hardness equation is applicable from 10 to 699 mg CaCO3 /L and should
not be applied outside of this range.
Guidelines for the protection of fresh water aquatic life at various hardness values
| Water hardness (mg/L as CaCO3) | Short-term exposure (mg Cl-/L) | Long-term exposure (mg Cl-/L) |
| Lower limit* | 403 | 50 |
| Soft (50) | 478 | 108 |
| Moderately hard (150) | 660 | 181 |
| Hard (300) | 810 | 251 |
| Upper limit** (796 and 699) | 1079 | 375 |
Notes: 1) the short-term hardness equation is applicable from 28 to 796 mg CaCO3 /L and should not be applied outside of this range;
2) the long-term hardness equation is applicable from 10 to 699 mg CaCO3 /L and should not be applied outside of this range.
* 403 mg Cl-/L is the lower limit short-term benchmark value that applies to waters less than 28 mg CaCO3/ L. 50 mg Cl-/L is the lower limit long-term WQG value that applies to waters less than 10 mg CaCO3/ L.
** 1079 mg Cl-/L is the upper limit short-term benchmark value that applies to waters greater than 796 mg CaCO3/ L. 375 mg Cl-/L is the upper limit long-term WQG value that applies to waters greater than 699 mg CaCO3/ L.
# 09-9192-50 / 13-AU-SGRC-10 / 15-SGRC-05 / 16-SGRC-03 / 17-SGRC-01 / 18-RRRC-05