Draft federal environmental quality guidelines - Benzene, toluene, ethylbenzene, xylene (BTEX)

Official title: Canadian Environmental Protection Act, 1999 - Federal Environmental Quality Guidelines - Benzene, Toluene, Ethylbenzene, Xylene (BTEX)

Environment and Climate Change Canada

June 2023

Introduction

Federal Environmental Quality Guidelines (FEQGs) describe acceptable quality of the ambient environment. They are based solely on the toxicological effects or hazards of specific substances or groups of substances. FEQGs serve three functions: first they can be an aid to prevent pollution by providing targets for acceptable environmental quality; second, they can assist in evaluating the significance of concentrations of chemical substances currently found in the environment (monitoring of water, sediment, soil and biological tissue); and third, they can serve as performance measures of the effectiveness of risk management activities. The use of FEQGs is voluntary unless prescribed in permits or other regulatory tools. Thus FEQGs, which apply to the ambient environment, are not effluent limits or “never-to-be-exceeded” values but may be used to derive effluent limits. The development of FEQGs is the responsibility of the Minister of Environment under the Canadian Environmental Protection Act, 1999 (CEPA) (Canada 1999). The intent is to develop FEQGs as an adjunct to risk assessment or risk management of priority chemicals identified in the Chemicals Management Plan (CMP) or other federal initiatives.

Where data permit, FEQGs are derived following Canadian Council of Ministers of the Environment (CCME) protocols. FEQGs are developed where there is a federal need for a guideline (for example, to support federal risk management or other monitoring activities) but where the CCME guidelines for the substance have not yet been developed or are not reasonably expected to be updated in the near future. For more information, please visit the Federal Environmental Quality Guidelines (FEQG) page.

This factsheet describes the FWQGs for the protection of aquatic life from adverse effects of benzene, toluene, ethylbenzene and xylene (BTEX) (Table 1). These FWQGs apply to both freshwater and marine environments. There are no pre-existing FWQGs for BTEX. Interim CCME guidelines for benzene, toluene and ethylbenzene exist (CCME 1999a,b,c) but they were developed following the CCME protocol (CCME 1991) that is no longer in use. No FEQGs for BTEX are developed for biological tissue, sediment or soil compartments since BTEX are not expected to accumulate in these environmental compartments.

Substance identity

BTEX is a group of monoaromatic hydrocarbons, consisting of benzene (C₆H₆; CAS Registry Number 71-43-2), toluene (C₇H₈; CAS RN 108-88-3), ethylbenzene (C₈H₁₀; CAS RN 100-41-4) and xylene (C₈H₁₀; CAS RN 1330-20-7). Xylene has three isomers commonly used in commercial products (o-xylene CAS RN 95-47-6, m-xylene CAS RN 108-38-3 and p-xylene CAS RN 106-42-3). BTEX compounds are flammable, clear and colourless liquids, naturally found in crude oil and petroleum products (Health Canada 2014). BTEX are volatile organic compounds (VOCs) that evaporate quickly into the atmosphere due to their high vapour pressures (Health Canada 2014). BTEX have been measured in oil sands process water (OSPW) in the northern Athabasca Basin in Alberta and are classified as substances of concern (Government of Alberta 2011; Mahaffey and Dubé 2016). Benzene and other VOCs are listed under the List of Toxic Substances in Schedule 1 of CEPA.

Sources and uses

In Canada, the sources of BTEX include both natural and anthropogenic. Natural sources of BTEX include petrogenic substances including coal as well as emissions from volcanoes and forest fires (Health Canada 2007, 2014; NTP 2016). BTEX are used in, and produced by, oil and natural gas operations, and are present in gasoline products (CAREX Canada 2020a,b). Anthropogenic sources of BTEX are usually produced from the synthesis of other chemical compounds or released to the environment during their use as industrial solvents. Benzene is used as a raw material in the production of other chemicals, including ethylbenzene to produce styrene, cumene to produce phenol and acetone, and cyclohexane to produce nylon and synthetic fibres (ATSDR 2007a; CAREX Canada 2020a; NTP 2016). It is also used in the manufacturing of products such as detergents, drugs, lubricants, rubbers and pesticides (ATSDR 2007a; CAREX Canada 2020a). Sources of benzene in surface water are typically from industrial effluents and atmospheric pollution (Health Canada 2009). Toluene is used in the synthesis of chemicals such as benzene and toluene diisocyanate. It is also used as a solvent in several products and processes, including adhesives, fingernail polish, lacquers, leather tanning, paint thinners, paints, printing and rubber (ATSDR 2017; Health Canada 2014). Ethylbenzene is primarily used in the production of styrene and polystyrene foam (Health Canada 2007, 2014). Ethylbenzene is also used as a solvent in consumer products, although this use is less prevalent than its use in styrene production (CAREX Canada 2020b; Health Canada 2007). Ethylbenzene can be found in consumer products such as adhesives, coatings dyes, perfumes, pharmaceuticals, plastics, rubber and varnishes (ATSDR 2007c; CAREX Canada 2020b; Health Canada 2007). Xylene is used primarily as a solvent in cleaning agents, paint thinners and varnishes (ATSDR 2007c; Health Canada 2014).

Ambient concentrations

BTEX compounds are typically not routinely tested in water monitoring programs due to their short retention times in surface water. Monitoring of BTEX in surface water is usually only performed around industrial operations that are likely to have elevated concentrations from effluent emissions. Even in areas of high industrial activity, the majority of water samples are found to have BTEX concentrations below detection limits.

Table 2 provides information on the concentration of BTEX compounds in surface waters in the Athabasca Region of Alberta, a region impacted by oil sands. The surface water monitoring data were collected between 2011 and 2017 from Alberta’s Regional Aquatics Monitoring Program (RAMP) for the Athabasca region (RAMP 2021). For computation of mean and median BTEX concentrations, values below method detection limits (MDL) were treated as half the detection limit. In addition, ECCC, HC (2016) provides additional ambient data for ethylbenzene in Canadian surface waters.

Source: Data from Regional Aquatics Monitoring Program (RAMP) (2021).
^a MDL = method detection limit.
^b m+p and o are xylene isomers.

Mode of action

Similar to most petroleum derived compounds, the mode of toxic action of BTEX is believed to be non- specifically acting narcosis (Hsieh et al. 2006; Modrzyński et al. 2019; Posthuma et al. 2019). BTEX are included in the target lipid model training set with over 200 other narcotic chemicals (McGrath et al. 2018). Narcotic chemicals accumulate in tissues and exert effects through non-specific interference with cell membranes (Hsieh et al. 2006; Li et al. 2013). Since a release of one of the BTEX chemicals alone is unlikely, it is important to note that when the compounds exist in a mixture the effects of the narcotic toxic action are by way of concentration addition. There is also some contradicting information regarding the potential for endocrine activity by BTEX, with some studies suggesting that BTEX have demonstrated endocrine activity (Kassotis et al. 2014; Robert et al. 2019) and others suggesting they lack endocrine activity (Mihaich and Borgert 2018). Most focus is on human health affects via exposure to air (Bolden et al. 2015) as opposed to aquatic life.

Fate, behaviour and partitioning in the environment

BTEX are cycled through the air, water and soil, fed by industrial activity and from natural sources. BTEX are rapidly biodegraded in the environment and are non-persistent in the aquatic environment (CCME 1999a,b,c; ECCC, HC 2016). Volatilization is the primary physical property of BTEX that determines the fate and persistence of these compounds in the environment. BTEX can stay in the atmosphere until they partition through photo-oxidation upon reaction with other substances in the air (Bandow et al. 1985). BTEX in surface water are likely to partition into the atmosphere through volatilization (ATSDR 2007a; Health Canada 2014). Benzene is the most soluble compound of the group but will not remain long in surface water due to its rapid volatilization half-life in surface water of ~5 hours (ATSDR 2007a). Toluene will remain in surface water for 5 hours to 16 days before it volatizes into the atmosphere, while ethylbenzene and xylene half-lives vary in turbulent and static water from 3.1 hours to 4.1 days (Health Canada 2014).

Water solubilities for benzene, toluene, ethylbenzene and xylene are 1780 mg/L, 515 mg/L, 152 mg/L and 175-198 mg/L, respectively at 20℃ (Headley et al. 2000). The log K_ow values in order of benzene, toluene, ethylbenzene and xylene are 1.99, 2.54, 3.03 and 3.09, respectively (McGrath et al. 2018). Due to the low log K_ow values of BTEX, bioaccumulation of BTEX in aquatic biota is expected to be minimal. In addition, there is insufficient evidence available to determine whether or not biomagnification of BTEX occurs up the aquatic food chain (Gossett et al. 1983). BTEX are subject to biodegradation in water and soil under aerobic conditions (ATSDR 2007a,b). BTEX compounds are very mobile in wet soil and are subject to leaching into groundwater which in turn can discharge to surface waters, but are otherwise readily volatile in dry soil (ATSDR 2007a,b,c).

Aquatic toxicity data

The datasets presented in this document and used for the species sensitivity distributions (SSDs) and acute to chronic ratios (ACRs) are based on the collection and evaluation of aquatic toxicity data published up to August 2021. A detailed review of studies was performed by ECCC following the CCME (2007) guidance for data quality. Determinants of test acceptability included, but were not limited to, exposure duration, documentation of the control response, the use of suitable biological endpoints and the inclusion of appropriate statistical analyses of the data collected in the study. Due to the volatility of these compounds, special consideration was given to whether concentrations were analytically measured, if concentrations were maintained throughout the test duration, if test vessels were designed to contain the chemicals (for exemple, sealed, air space eliminated, flow-through system), and if standard procedures were followed. Studies were classified as either primary, secondary or unacceptable using CCME (2007) as guidance.

The Target Lipid Model (TLM) training set (McGrath et al. 2018) includes over 1000 endpoints from many studies including those for BTEX. The entire training set was not independently assessed here but many of the same BTEX papers were included in the SSD dataset and therefore were assessed by ECCC. Before inclusion into the TLM, studies were evaluated by the authors according to the Klimisch et al. (1997) scoring system and Organisation for Economic Co-operation and Development (OECD) (2000) guidelines, and only data deemed reliable were accepted (McGrath et al. 2018).

There is some uncertainty with the datasets since many of the lack important test details, especially for volatile chemicals like BTEX. Studies were deemed unacceptable for guideline derivation when the uncertainty surrounding maintaining test concentrations and appropriate test design was deemed too large.  Toxicity data considered for developing the FWQGs are presented in the Appendix including reasoning for data quality classification. Endpoints for chronic exposure durations are scarce for all BTEX chemicals which is recognized as an uncertainty. Toxicity endpoints within and across chemicals typically span a narrow range, likely due to the chemicals sharing similar physical and chemical properties including a non-specific acting narcotic mode of action. Although the endpoints did not range widely, it appears that pelagic invertebrates, specifically cladoceran (that is, water fleas), and amphibians are most sensitive to BTEX.

Acceptable chronic amphibian data were available for two amphibian species (Black at al. 1982; Kennedy 2006) for benzene, toluene and xylene but no data were available for ethylbenzene or acute exposures. Both studies had comparable endpoints for Lithobates pipiens (leopard frog) for benzene and xylene which were more sensitive than the rest of the species in the datasets. Black et al. (1982) also tested the toxicity of benzene to Ambystoma gracile (northwestern salamander) and endpoints were aligned with L. pipiens. The amphibian and Oncorhynchus mykiss (rainbow trout) endpoints from Black et al. (1982) for toluene were considerably lower than the rest of the endpoints in the guideline dataset. Kennedy (2006) appears to support the O. mykiss value from Black et al. (1982) with similar endpoints values. However, there are no endpoints from Kennedy (2006) or other studies to compare against the toluene amphibian data. Black et al. (1982) was included in the development of the CCME interim guidelines (CCME 1999a,b) and was deemed to be acceptable based on data quality assessment for guideline development for BTEX in this document. Black et al. (1982) and Kennedy (2006) data were not included in the TLM for unknown reasons. However, over one thousand endpoints and 72 individual ACRs from over 200 chemicals are included in the TLM so it is uncertain how great an influence including these endpoints would make to the TLM estimates. Additional research on the toxicity to amphibians from BTEX is needed, specifically chronic exposure including full-life cycle reproductive tests to consider potential endocrine affects (Robert et al. 2019).

Federal water quality guideline derivation

Federal Water Quality Guidelines (FWQGs) are preferably developed using the CCME (2007) protocol, however they can be developed following other methods when data requirements are not met. In the case of BTEX, there were sufficient acute toxicity data to develop Type A benchmarks but insufficient chronic toxicity data to develop Type A or B long-term guidelines (CCME 2007). Therefore, for the long-term guidelines, an ACR was used to extrapolate endpoints from short-term to long-term and then used with available long-term data to meet the minimum data requirements. The most sensitive and preferred endpoint (or geometric mean) was then selected for each species following CCME (2007). A geometric mean was calculated where multiple comparable endpoints were available for the same species, effect, life stage and exposure duration. Where no algae or aquatic plant endpoints were available for acute exposures, typically defined as less than or equal to 24 hours, median endpoints for 48-hours were included in the acute datasets instead as per CCME (2007). Some EC10 endpoints were calculated by ECCC using the USEPA toxicity relationship analysis program (TRAP v. 1.3) (USEPA 2015) and are identified in the Appendix under “additional notes”.

Species sensitivity distributions

The R package (R version 4.03) ‘ssdtools’ (ssdtools version 1.0.2) as well as the corresponding user friendly web application ‘shinyssdtools’ (shinyssdtools version 0.1.1) were used to create SSDs from the datasets (Dalgarno 2018; Thorley and Schwarz 2018). The package fits several cumulative distribution functions (CDFs) (log-normal, log-logistic, gamma, log-Gumbel, Weibull, log-normal mixture) to the data using maximum likelihood estimation (MLE) as the regression method. The package then uses Akaike information criterion (AICc) to weight each model, representing how well each fit the data relative to the others. The best predictive model is that with the lowest AICc (indicated by the model with a delta value of 0). The distributions that successfully fit the data with delta values <7 are averaged based on the respective weights. HC_p estimates with 95% confidence intervals are calculated using parametric bootstrapping (10000 iterations). The hazard concentration at 5% (herein referred to as HC₅) output is then used as the benchmark or guideline value if deemed protective. The full R script and results are available in the Appendix. See Fox et al. (2020) and Thorley and Schwarz (2018) for more information on the approach.

Target lipid model

Since BTEX are considered nonpolar narcotics and the datasets were not robust, the target lipid model (TLM) was used as an additional line of evidence to support and, in some cases, establish the acute benchmarks and long-term guidelines developed for BTEX. The TLM is a quantitative structure-activity relationship (QSAR) developed for nonpolar narcotics based on the premise of critical body burden theory (McCarty et al. 1991, 1992; Di Toro et al. 2000; McGrath et al. 2018). The model predicts no-effect concentrations in water based on a chemical’s K_OW, represented by the lower confidence level HC₅ (herein referred to the TLM-based HC₅) which was found to be protective of 95% of species in the model dataset (McGrath et al. 2018). Computer (EPI Suite Ver 4.11) estimated log-K_OW inputs for benzene (1.99), toluene (2.54), ethylbenzene (3.03) and xylene (3.09) from McGrath et al. (2018) were used for all TLM equations. The following TLM equation derives chronic HC₅ values for Type I narcotic chemicals (with log K_OW <6.5) (McGrath et al. 2018):

Equation 1:

$Chronic Log(H{C}_5)=E[m]log{K}_{OW}+E[log{(C}_L^{*})]+ \Delta c – E[log(ACR)]– K_Z \sqrt{V[m]{log{(K}_{OW})}^2+V[log{(C}_L^{*})]+V[log(ACR)]+2log{(K}_{OW})[Cov(m,log{(C}_L^{*})]}$

where, the universal narcosis slope is E[m] = -0.940 with a variance of V[m]=0.000225, the log mean value of 79 critical target lipid body burdens (CTLBBs) is E[log(C*_L)]=1.85 with a variance of V[log(C*_L)]=0.135, ACR= 5.22, log mean acute to chronic ratio (ACR) is E[log(ACR)]=0.718 with a variance of V[log(ACR)]=0.149, the covariance between slope and log CTLBB Cov(m,log(C*_L))= -0.0079, and the 95% confidence sample size-dependent extrapolation factor k_Z=2.396 (McGrath et al 2018).

The species geometric average of 65 definitive ACRs is 5.22 and is used in the TLM equation above. The TLM equation for acute effect concentrations in water (that is, acute HC₅ values) for Type I narcotic chemicals (with log K_OW <6.5) is the same as Equation 1 but with the ACR terms removed (McGrath et al. 2018):

Equation 2:

$Acute Log(H{C}_5)=E[m]log{K}_{OW}+E[log{(C}_L^{*})]+ \Delta c – K_Z \sqrt{V[m]{log{(K}_{OW})}^2+V[log{(C}_L^{*})]+2log{(K}_{OW})[Cov(m,log{(C}_L^{*})]}$

The TLM has been published in several peer-reviewed journal articles and includes data for a large number of species and narcotics. The model has been validated for mono-aromatic hydrocarbons like BTEX and BTEX toxicity data are included in the training set. The TLM represents an opportunity to derive guidelines for nonpolar narcotics with limited data and was used to develop an aquatic biota tissue guideline for siloxane D4 (ECCC 2022).

To be conservative, the lower value determined among the two methods (that is, TLM and SSD) for each unique toxicity dataset was set as the guideline value. The BTEX datasets were compiled with mostly freshwater species but some marine species were also included as indicated. This is aligned with the TLM which also includes species from both environments since nonpolar narcotics are not expected to vary in toxicity in the two environments. Therefore the BTEX FEQGs can be applied to both freshwater and marine environments.

Acute to chronic ratios

Acute SSDs were created for all four BTEX chemicals using acute toxicity data. Since insufficient data existed for full chronic SSDs, the chronic data were supplemented with transformed acute data using an ACR. Paired acute and chronic endpoints from the same study were used to determine ACRs and then a geometric mean was taken of all available ACRs to be used for all BTEX chemicals.

For benzene, paired acute and chronic endpoints from the same study were available for two species: Ceriodaphnia dubia (water flea, invertebrate) and Pimephales promelas (fathead minnow, fish). An ACR of 3.4, was derived by dividing the C. dubia acute lethal endpoint (48-h LC50=17.3 mg/L) by the chronic maximum acceptable toxicity concentration (7-d MATC=5.14 mg/L) for reproduction, both obtained from Niederlehner et al. (1998). The second ACR, 1.5, was derived by dividing the P. promelas acute lethal endpoint (96 h LC50=15.59 mg/L) by the chronic 25% effect concertation (7-d EC25=10.6 mg/L) for biomass, both obtained from Marchini et al. (1992).

There were paired acute and chronic endpoints for toluene from the same study for three species: C. dubia, P. promelas and Oncorhynchus kisutch (coho salmon, fish). An ACR of 2.6, was derived by dividing the C. dubia acute lethal endpoint (48-h LC50=3.78 mg/L) by the chronic maximum acceptable toxicity concentration (7-d MATC= 1.43 mg/L) for reproduction, both obtained from Niederlehner et al. (1998). The acute lethal endpoint (96 h LC50=17.03 mg/L) was divided by the chronic 25% effect concertation (7-d EC25=6.53 mg/L) for biomass, both obtained for P. promelas from Marchini et al. (1992), resulting in an ACR of 2.6. The acute lethal endpoint (96-h LC50=6.3) was divided by the chronic maximum acceptable toxicity concentration (40-d MATC=2.28 mg/L), both obtained for O. kisutch from Moles (1981), resulting in an ACR of 2.8.

There were no paired acute and chronic endpoints for ethylbenzene or xylene. Therefore, a geometric mean was taken of the calculated ACRs of 3.4, 1.5, 2.6, 2.7 and 2.8 resulting in an average ACR of 2.5 to be used for all BTEX chemicals for transforming acute data for chronic SSDs. 

Benzene

Short-term benchmark

A total of 13 endpoints for 13 species (5 fish, 6 invertebrates, 2 aquatic plants/algae) from 9 studies were included in the SSD dataset and are summarized in Table 3. Three marine species and 10 freshwater species were included in the SSD dataset. O. mykiss (fish) was the most sensitive species in the dataset with a median effect concentration of 5.9 mg/L. Artemia sp. (invertebrate) was the least sensitive species in the dataset with a median effect concentration of 127.3 mg/L. Figure 1 shows the acute SSD for benzene, with a resulting HC₅ of 6 (4.2-10) mg/L. The TLM-based acute HC₅ of 11.8 mg/L was calculated using Equation 2 and 1.99 for log K_OW. The SSD- (6 mg/L) and TLM-based (11.8 mg/L) acute HC₅s were in good agreement, being within a factor of 2 of each other.

Long-term guideline

Very few chronic data were available and the dataset did not meet CCME requirements for long-term Type A guideline development, lacking at least two fish and two invertebrate endpoints from acceptable studies. Combining the chronic data and the transformed acute data using the ACR of 2.5, a SSD was derived following the procedures described previously. Table 4 includes chronic data for 6 species and transformed acute data for 10 species. Figure 2 shows the chronic SSD for benzene, with a resulting HC₅ of 0.59 (0.22-2.0) mg/L. Equation 1 was used to derive a TLM-based HC₅ of 0.84 mg/L. The SSD- and TLM-based HC₅s are in good agreement and well within a factor of two.

The two amphibians (L. pipiens and A. gracile) showed higher sensitivity compared to the rest of the dataset (as discussed previously) falling just below both the SSD HC₅ and TLM estimate. The available fish, invertebrate and algae endpoints are above both HC₅ estimates, as well as all acute endpoints in the acute dataset. The L. pipiens (leopard frog) Rocky Mountain population is designated as endangered and the Western Boreal/Prairie population is designated as a population of concern under Schedule 1 of the Species at Risk Act (SARA SC 2002, c.29). The L. pipiens eastern population and A. gracile are not at risk. Since the L. pipiens LC10 of 0.55 mg/L, and another L. pipiens LC20 (subchronic) value of 0.4 mg/L (Kennedy 2006) not in the SSD dataset are below the SSD HC5 of 0.59 mg/L, the CCME protection clause (CCME 2007) should be considered. The protection clause states that if an acceptable no-effect or low-effect level endpoint for a species at risk in Canada is lower than the proposed long-term guideline then that endpoint becomes the recommended guideline. However, some uncertainties exist regarding the reliability of Black et al. (1982) endpoints. Because of this uncertainty and since the SSD-based HC₅ is negligibly higher than the lowest SSD endpoint, the SSD estimate of 0.59 mg/L was adopted as the long-term guideline for benzene.

^a Acute effect concentrations were transformed using an ACR of 2.5. Chronic concentrations were left untransformed.
^b When the endpoint effect is not lethality (that is, LCx), the effect is noted in brackets.
^c Calculated by ECCC by taking the geometric mean of the NOEC (2.97 mg/L) and LOEC (8.9 mg/L) from study.

^a Values are the lower values and are adopted as the benchmark and FEQG.

Toluene

Short-term benchmark

A total of 14 endpoints for 14 species (5 fish, 7 invertebrates, 2 aquatic plants/algae) from 11 studies were included in the SSD dataset and are summarized in Table 6. Ceriodaphnia dubia (invertebrate) was the most sensitive species in the dataset with a median effect concentration of 3.78 mg/L. Scenedesmus subspicatus (algae) was the least sensitive species in the dataset with a median effect concentration of 125 mg/L. Two marine species and 12 freshwater species were included in the SSD dataset. Figure 3 shows the acute SSD for toluene with a resulting HC₅ of 3.0 (1.7-6.9) mg/L. The TLM-based acute HC₅ of 4.6 mg/L was calculated using Equation 2. The SSD- (3.0 mg/L) and TLM-based (4.6 mg/L) acute HC₅ are in good agreement, being within a factor of 2 of each other.

Long-term guideline

Limited chronic data were available and the dataset did not meet CCME requirements for guideline development, lacking at least two invertebrate endpoints from acceptable studies. Combining the few chronic data and the transformed acute data using the ACR of 2.5, an SSD was derived following the procedures described previously. Table 7 includes chronic data for 8 species and transformed acute data for 9 species. Figure 4 shows the chronic SSD for toluene, with a resulting HC₅ of 0.03 (0.002-0.55) mg/L.

The TLM-based chronic HC₅ of 0.31 mg/L was calculated using Equation 1. The SSD-based HC₅ (0.03 mg/L) is 10-fold lower than the TLM-based HC₅ (0.31 mg/L). SSD endpoints for two species (L. pipiens and O. mykiss) fall below the SSD HC₅ and three species (additionally, A. gracile) fall below the TLM HC₅. Both L. pipiens (Rocky Mountain and the Western Boreal/Prairie) and O. mykiss populations (Athabasca River) are listed under Schedule 1 of the Species at Risk Act (SARA S.C.2002,c.29 ) and therefore the CCME protection clause (CCME 2007) should be considered. However, as was the case with benzene, there are some uncertainties with this dataset due to the lack of robust chronic data and because the three most sensitive species are far lower than the rest of the dataset. The lowest effect concentration of 0.006 mg/L is over 50-fold lower than what is estimated by the TLM (0.31 mg/L) and 5-fold lower than the SSD HC₅ (0.03 mg/L). Therefore, in order to balance the desire to be conservative and to acknowledge the uncertainties with and the reliability of the available data, the SSD estimate of 0.03 mg/L was adopted as the long-term guideline for toluene. Additional consideration may be warranted in sites where amphibians and rainbow trout are present. 

^a Acute effect concentrations for fish and invertebrates were transformed using the ACR of 2.5 and chronic concentrations were left untransformed.
^b When the endpoint effect is not lethality (that is, LCx), the effect is noted in brackets.
^c Calculated by ECCC by taking the geometric mean of the NOEC and LOEC.

^a Values are the lower values and are adopted as the benchmark and FEQG.

Ethylbenzene

Short-term benchmark

A total of 12 endpoints for 12 species (3 fish, 7 invertebrates, 2 aquatic plants/algae) from 7 studies were included in the SSD dataset and are summarized in Table 9. Daphnia magna (invertebrate) was the most sensitive species in the dataset with a median effect concentration of 2.12 mg/L. Chironomus plumosus (invertebrate) was the least sensitive species in the dataset with a median effect concentration of 37.8 mg/L. Three marine species and nine freshwater species were included in the dataset. Figure 5 shows the acute SSD for ethylbenzene with a resulting HC₅ of 1.9 (1.2-3.5) mg/L.

The TLM-based acute HC₅ was calculated using Equation 2 and 3.03 for log K_OW. This resulted in a TLM-based HC₅ of 1.9 mg/L. Therefore, both estimates corroborate each other, providing strong support for the resulting benchmark.

Long-term guideline

Very few chronic data were available and the dataset did not meet CCME requirements for guideline development, missing data for fish and at least one other invertebrate. Combining the limited chronic data and the transformed acute data using the ACR of 2.5, an SSD was derived following the procedures described previously. Table 10 includes chronic data for 4 species and transformed acute data for 9 species. Figure 6 shows the chronic SSD for ethylbenzene, with a resulting HC₅ of 0.37 (0.1-1.2) mg/L. Equation 1 was used to derive a TLM-based HC₅ which resulted in a value of 0.13 mg/L. All available chronic and acute data were above the TLM-based estimate however one endpoint, the 21-day MATC (reproduction) of 0.2 mg/L for D. magna (Kennedy 2006) did fall below the SSD HC₅. Daphnia magna is not a species at risk in Canada and the endpoint is for reproduction, not mortality, and therefore the CCME (2007) protection clause is not invoked. However, due to the limitations of the chronic SSD and because the TLM estimate is lower than the SSD HC₅, the TLM-based HC₅ estimate of 0.13 mg/L was adopted as the guideline for ethylbenzene.

^a Acute effect concentrations for fish and invertebrates were transformed using the ACR of 2.5 and chronic concentrations were left untransformed.
^b When the endpoint effect is not lethality (that is, LCx), the effect is noted in brackets.
^c Calculated by ECCC by taking the geometric mean of the NOEC and LOEC.

^a Values are the lower values and are adopted as the benchmark and FEQG.

Xylene

Short-term benchmark

A total of 23 endpoints for 12 species (6 fish and 6 invertebrates) from 7 studies were included in the SSD dataset and are summarized in Table 12. Some studies measured the three isomers (o-, m- and p-) separately while others measured them as a total. The available data do not suggest that one isomer is more toxic than the others, which is also supported by the narcotic mode of action. In addition, sufficient data were not available to create separate SSDs for all three isomers. Therefore, all isomers were grouped together in the same dataset. Where comparable endpoints exist for the same species, a geometric mean was calculated. For example, a geometric mean of endpoints for o-, m- and p- isomers were taken in several instances and used in the SSD. Ceriodaphnia dubia (invertebrate) was the most sensitive species in the dataset with a median effect concentration of 2.44 mg/L. Chironomus plumosus (invertebrate) was the least sensitive species in the dataset with a median effect concentration of 42 mg/L. One marine species and 11 freshwater species were included in the dataset. Figure 7 shows the acute SSD for xylene with a resulting HC₅ of 2.2 (1.2-4.6) mg/L.

The TLM-based acute HC₅ was calculated using Equation 2 and 3.09 for log K_OW. This results in a TLM-based HC₅ of 1.7 mg/L. The SSD- (2.2 mg/L) and TLM-based (1.7 mg/L) short-term HC₅s are in good agreement, being within a factor of 2 of each other.

Long-term guideline

Very few chronic data were available and the chronic dataset did not meet CCME requirements for guideline development, missing at least two fish and two invertebrate species. Combining the chronic data and the transformed acute data using the ACR of 2.5, an SSD was derived following the procedures described previously. Table 13 includes chronic data for 4 species and transformed acute data for 11 species. Figure 8 shows the chronic SSD for xylene, with a resulting HC₅ of 0.56 (0.21-1.5) mg/L. Equation 1 was used to derive a TLM-based HC₅ which resulted in a value of 0.12 mg/L. All available chronic and acute data are above the TLM-based estimate however one endpoint, the 6-day LC20 (embryo-larval mortality) of 0.31 mg/L for L. pipiens (amphibian) (Kennedy 2006) does fall below the SSD HC₅. Therefore, as was done for benzene and toluene, the protection clause (CCME 2007) was considered. However, the lowest value (0.31 mg/L) is still 2.5 fold higher than the TLM estimate. To be conservative and consistent, the TLM estimate was adopted as the guideline value for xylene.

^a Acute effect concentrations for fish and invertebrates were transformed using the ACR of 2.5 and chronic concentrations were left untransformed.
^b When the endpoint effect is not lethality (that is, LCx), the effect is noted in brackets. 

^a Values are the lower values and are adopted as the benchmark and FEQG.

Federal water quality guidelines

A short-term benchmark concentration and the long-term FWQG provide guidance for both acute and chronic exposures, respectively. The short-term exposure value is intended to protect most species, not individuals,

against lethality during severe but transient events such as spills or inappropriate use/disposal of the substance in question. Long-term guidelines are intended to protect the most sensitive species and life stages indefinitely. Although BTEX are not persistent, aquatic life may be chronically exposed to a substance because of gradual release from soils/sediments, gradual entry through groundwater/runoff and effluents from industrial processes. The short-term benchmark concentrations and FWQGs for BTEX are tools for the assessment and interpretation of BTEX monitoring data in water.

CCME type A or Type B guidelines could not be derived for BTEX because all required chronic toxicity data were not available. However, appropriate data were available to calculate an ACR to fulfill minimum data requirements to derive SSDs. Since BTEX are all considered nonpolar narcotics, the TLM was used to corroborate the SSD-based guideline or derive the FEQGs when the estimate was more conservative.

The final short-term benchmarks and long-term guidelines for BTEX compounds are the lower of the SSD- and TLM-based estimates. The federal short-term benchmarks and long-term guidelines for BTEX compounds were rounded to two significant figures and are presented in Table 1. These FWQGs apply to both freshwater and marine environments.

References

Abernethy, S., A. Bobra, W. Shiu, P. Wells and D. Mackay. 1986. Acute lethal toxicity of hydrocarbons and chlorinated hydrocarbons to two planktonic crustaceans: The key role of organism-water partitioning. Aquatic Toxicol. 8: 163-174.

[ATSDR] Agency for Toxic Substances and Disease Registry. 2007a. Toxicological Profile for Benzene. U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA.

[ATSDR] Agency for Toxic Substances and Disease Registry. 2007b. Toxicological Profile for Ethylbenzene. U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA.

 [ATSDR] Agency for Toxic Substances and Disease Registry. 2007c. Toxicological Profile for Xylenes. U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA.

[ATSDR] Agency for Toxic Substances and Disease Registry. 2017. Toxicological profile for Toluene. U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA.

Bandow, H., N. Washida and H. Akimoto. 1985. Ring-cleavage reactions of aromatic hydrocarbons studied by FT-IR spectroscopy: I. Photooxidation of toluene and benzene in the NOx-air system. Bul. Chem. Soc. Japan 58: 2531-2540. 

Black, J.A., W.J. Birge, W.E. McDonnell, A.G. Westerman, B.A. Ramey and D.M. Bruser. 1982. The aquatic toxicity of organic compounds to embryo-larval stages of fish and amphibians. University of Kentucky Water Resources Research Institute, Report No. 133.

Bolden, A.L., C.F. Kwiatkowski and T. Colborn. 2015. New Look at BTEX: Are Ambient Levels a Problem?  Environ. Sci. Tech.  49: 5261-5276

Brenniman, G., R. Hartung and W.J. Weber. 1976. A continuous flow bioassay method to evaluate the effects of outboard motor exhausts and selected aromatic toxicants on fish. Water Research 10: 165-169.

[CCME] Canadian Council of Ministers of the Environment.  1991.  Protocol for the derivation of water quality guidelines for the protection of aquatic life.  Canadian Council of Ministers of the Environment, Canadian Water Quality Guidelines.

[CCME] Canadian Council of Ministers of the Environment. 1999a. Canadian Water Quality Guidelines for the Protection of Aquatic Life: Benzene. Canadian Environmental Quality Guidelines, 2005 Update

[CCME] Canadian Council of Ministers of the Environment. 1999b. Canadian water quality guidelines for the protection of aquatic life: Toluene. In: Canadian environmental quality guidelines, 1999, Canadian Council of Ministers of the Environment, Winnipeg.

[CCME] Canadian Council of Ministers of the Environment. 1999c. Canadian water quality guidelines for the protection of aquatic life: Ethylbenzene. In: Canadian environmental quality guidelines, 1999, Canadian Council of Ministers of the Environment, Winnipeg.

[CCME] Canadian Council of Ministers of the Environment. 2007. A Protocol for the Derivation of Water Quality Guidelines for the Protection of Aquatic Life. In: Canadian Environmental Quality Guidelines, 1999, Canadian Council of Ministers of the Environment, Winnipeg.

CAREX Canada. 2020a. Benzene Profile. [accessed 7 June 2022]

CAREX Canada. 2020b. Ethylbenzene Profile. [accessed 7 June 2022]

Dalgarno, S. 2018. ssdtools: A Shiny Web App to Analyse Species Sensitivity Distributions. Prepared by Poisson Consulting for the Ministry of the Environment, British Columbia. [accessed 16 December 2020]

Di Toro, D.M., J.A. McGrath and D.J. Hansen. 2000. Technical basis for narcotic chemicals and PAH criteria. I. Water and tissue. Environ. Toxicol. Chem. 19: 1951-1970.

[ECCC] Environment and Climate Change Canada. 2022. Federal Environmental Quality Guidelines for Siloxane D4 (Octamethylcyclotetrasiloxane). National Guidelines and Standards Office, Environment and Climate Change Canada, Gatineau, QC.

[ECCC, HC] Environment and Climate Change Canada, Health Canada. 2016. Screening assessment: Ethylbenzene. Ottawa (ON): Government of Canada.

Fox, D.R., R.A. van Dam, R. Fisher, G.E. Batley, A.R. Tillmanns, J. Thorley, C.J. Schwarz, D.J. Spry and K. McTavish. 2021. Recent developments in species sensitivity distribution modeling. Environ. Toxicol. Chem. 40: 293-308.

Furay, V. and S. Smith. 1995. Toxicity and QSAR of chlorobenzenes in two species of benthic flatfish, flounder (Platichthys flesus L.) and sole (Solea solea L.). Bull. Environ. Cont. Toxicol. 54: 36-42.

Galassi, S., M. Mingazzini, L. Vigano, D. Cesareo and M.L. Tosato. 1988. Approaches to modelling toxic responses of aquatic organisms to aromatic hydrocarbons. Ecotox. Environ. Saf. 16: 158-169.

Gossett, R.W., D.A. Brown and D.R. Young. 1983. Predicting the bioaccumulation of organic compounds in marine organisms using octanol/water partition coefficients. Marine Poll. Bull. 14: 387-392.

Government of Alberta. 2011. Draft Lower Athabasca integrated region plan 2011-2021. No. 1/511. Edmonton.

[GC] Government of Canada. 1999. Canadian Environmental Protection Act, 1999. S.C., 1999, c. 33, Canada Gazette. Part III, vol. 22, no. 3. Published by the Minister of Justice, Government of Canada.

Headley, J.V., S.L. Barbour, R.W Thring and Y. Gong. 2000. An evaluation of the ideality of benzene, toluene, ethylbenzene, and xylene on activity coefficients in gas condensate and the implications for dissolution in groundwater. Can. Wat. Res. 25:1 67-79.

Health Canada. 2007. Ethylbenzene and Health. Health Canada. (Catalogue No. H128-1/07-496-3E).

Health Canada. 2009. Guidelines for Canadian Drinking Water Quality: Guideline Technical Document—Benzene. Water, Air and Climate Change Bureau, Healthy Environments and Consumer Safety Branch, Health Canada, Ottawa, ON. (Catalogue No. H128-1/09-589E).

Health Canada. 2014. Guidelines for Canadian Drinking Water Quality: Guideline Technical Document — Toluene, Ethylbenzene and Xylenes. Water and Air Quality Bureau, Healthy Environments and Consumer Safety Branch, Health Canada, Ottawa, Ontario. (Catalogue No. H144-20/2015E-PDF).

Herman, D.C., W.E. Inniss and C.I. Mayfield. 1990. Impact of volatile aromatic hydrocarbons, alone and in combination, on growth of the freshwater alga Selenastrum capricornutum, Aquatic Toxicol. 18: 87-100.

Holcombe, P. and H. Sulaiman. 1987. Simultaneous multiple species testing: acute toxicity of 13 chemicals to 12 diverse freshwater amphibian, fish, and invertebrate families. Arch. Environ. Cont. Toxicol. 16: 697-709.

Hsieh, S., K. Tsai and C. Chen. 2006. The combined toxic effects of nonpolar narcotic chemicals to Pseudokirchneriella subcapitata. Wat. Res. 40: 1957-1964.

Kassotis C.D., D.E. Tillitt, J.W. Davis, A.M. Hormann and S.C. Nagel. 2014. Estrogen and androgen receptor activities of hydraulic fracturing chemicals and surface and ground water in a drilling-dense region. Endocrinology. 155: 897-907. Erratum in: 2015 Endocrinology 156: 4374.

Klimisch H.J., M. Andreae and U. Tillmann. 1997. A systematic approach for evaluating the quality of experimental toxicological and ecotoxicological data. Regul. Toxicol. Pharmacol 25: 1-5.

Kuhn, R. and M. Pattard. 1990. Results of the harmful effects of water pollutants to green algae (Scenedesmus subspicatus) in the cell multiplication inhibition test.  Water Res. 24: 31-38.

Li, X., Q. Zhou, Y. Luo, G. Yang and T. Zhou. 2013. Joint action and lethal levels of toluene, ethylbenzene, and xylene on midge (Chironomus plumosus). Environ. Sci. Pollut. Res. 20: 957-966.

Mahaffey, A. and M. Dubé. 2016. Review of the composition and toxicity of oil sands process-affected water. Environ. Rev. 25: 97-114.

Marchini, S., M.L. Tosato, T.J. Norberg-King, D.E. Hammermeister and M.D. Hoglund. 1992. Lethal and sublethal toxicity of benzene derivatives to the fathead minnow, using a short-term test. Environ. Toxicol. Chem. 11: 198-195.

Marzio, W.D. and M.E. Saenz. 2006. QSARs for aromatic hydrocarbons at several trophic levels. Environ. Toxicol. 21: 118-124.

Masten, L.W., R.L. Boeri and J.D. Walker. 1994. Strategies employed to determine the acute toxicity of ethyl benzene, a highly volatile, poorly water-soluble chemical. Ecotoxicol. Environ. Saf. 27: 335-348.

McCarty, L.S., D. Mackay, A.D. Smith, G.W. Ozburn and G.D. Dixon. 1991. Interpreting aquatic toxicity QSARs: the significance of toxicant body residues at the pharmacologic endpoint.  Sci. Total Environ. 109/110: 515-525.

McCarty, L.S., D. Mackay, A.D. Smith, G.W. Ozburn and G.D. Dixon. 1992. Residue-based interpretation of toxicity and bioconcentration QSARs from aquatic bioassays: Neutral narcotic organics.  Environ. Toxicol. Chem. 11: 917-930.

McGrath, J.A., C.J. Fanelli, D.M. Di Toro, T.F. Parkerton, A.D. Redman, L.M. Paumen, M. Comber, C.V. Eadsforth and K. den Haan. 2018. Re-evaluation of Target Lipid Model-derived HC5 predictions for hydrocarbons. Environ. Toxicol. Chem. 37: 1579-1593.

Mihaich, E.M. and C.J. Borgert. 2018 Hypothesis-driven weight-of-evidence analysis for the endocrine disruption potential of benzene. Regul. Toxicol. Pharmacol. 100: 7-15.

Modrzyński, J.J., J.H. Christensen and K.K. Brandt. 2019. Evaluation of dimethyl sulfoxide (DMSO) as a co-solvent for toxicity testing of hydrophobic organic compounds. Ecotoxicol. 28: 1136–1141.

Moles, A., S. Bates, D. Rice and S. Korn. 1981. Reduced growth of coho salmon fry exposed to two petroleum components, toluene and naphthalene, in fresh water. Trans. Am. Fish. Soc. 110: 430-436.

Niederlehner, B., J. Cairns and E.P. Smith. 1998. Modeling acute and chronic toxicity of nonpolar narcotic chemicals and mixtures to Ceriodaphnia dubia. Ecotoxicol. Environ. Saf. 39: 136-146.

[NTP] National Toxicology Program. 2016. Report on Carcinogens, Fourteenth Edition. U.S. Department of Health and Human Services, Public Health Service, Research Triangle Park, NC.

[OECD] Organisation for Economic Co-operation and Development. 2000. Guidance document on aquatic toxicity testing of difficult substances and mixtures. Series on Testing and Assessment, No. 23. ENV/JM/MONO(2000)6. Paris, France.

Philibert, D., T. Parkerton, T. Marteinson and B. de Joudan. 2021. Assessing the toxicity of individual aromatic compounds and mixtures to American lobster (Homarus americanus) larvae using a passive dosing system. Environ. Toxicol. Chem. 40: 1379-1388.

Posthuma, L., J. van Gils, M.C. Zijp, D. van de Meent and D. de Zwart. 2019 Species sensitivity distributions for use in environmental protection, assessment, and management of aquatic ecosystems for 12 386 chemicals. (Supporting Information) Environ. Toxicol. Chem. 38: 905-917.

[RAMP] Regional Aquatics Monitoring Program. 2021. RAMP Alberta Monitoring Database. [accessed 6 July 2021]

Robert J, C.C. McGuire, S. Nagel, B.P. Lawrence and F.J. Andino. 2019. Developmental exposure to chemicals associated with unconventional oil and gas extraction alters immune homeostasis and viral immunity of the amphibian Xenopus. Sci. Total Environ. 671: 644-654.

Rose, R.M., M. StJ. Warne and R.P. Lim. 1998. Quantitative structure–activity relationships and volume fraction analysis for nonpolar narcotic chemicals to the Australian cladoceran Ceriodaphnia cf. dubia. Arch. Environ. Contam. Toxicol. 34: 248-252.

[SARA] Species at Risk Act, SC 2002, c 29, Species search. [accessed 2022-07-03]

Thorley, J. and C. Schwarz. 2018. ssdtools: An R package to fit Species Sensitivity Distributions. J. Open Source Softw. 3: 1082.

Trucco, R., F.R. Engelhardt and B. Stacey. 1983. Toxicity, accumulation and clearance of aromatic hydrocarbons in Daphnia pulex. Environ. Pollut. 31: 191-202.

Tsai, K. and C. Chen. 2007. An algal toxicity database of organic toxicants derived by a closed system technique. Environ. Toxicol. Chem. 26: 1931-1939.

USEPA. 2015. Toxicity Relationship Analysis Program (TRAP) Version 1.30a. Mid-Continent Ecology Division, 6201 Congdon Blvd., Duluth, MN 55804.

Wong, P.T.S., Y.K. Chau, J.S. Rhaney and M. Docker. 1984.  Relationship between water solubility of chlorobenzenes and their effects on a freshwater green alga.  Chemosphere 13: 991-996.

List of Acronyms and Abbreviations