Rapid screening of substances of lower ecological concern: assessment

Assessment of toxic under paragraph 64(a) of CEPA 1999

In this section, an overview of the results obtained at each step of rapid screening of the substances covered under this assessment is provided. A spreadsheet that presents the detailed results for each substance at each step of the approach is available (Environment Canada, 2007b). A simplified spreadsheet, that provides only the outcome of each step of the approach for each of the substances, is also available (Environment Canada, 2007c).

Step 1 - Assessment categories

The first step of the approach identified substances that fall into one of the chemical structure categories considered as higher priorities for assessment - based on substances that were identified as PBiT under the categorization criteria. There were four substances undergoing rapid screening that were found to be structurally similar to PBiT substances on the DSL. Identification of these similar PiT and BiT substances during rapid screening provides a safety net to ensure that substances that may have fallen just short of meeting the P and B and iT categorization criteria are prioritized for assessment.

Step 2 - Exposure estimation

Generic aquatic scenarios

Two generic exposure scenarios are used to evaluate the potential for releases of a substance to the aquatic environment to cause harm. The first scenario (industrial point-source) considers the release of 5% of the substance from an industrial facility that is involved in manufacturing the substance and/or using it in the preparation of products. The second scenario (residential releases to municipal wastewater) considers the down-the-drain release of 100% of the substance that is contained in a consumer product, through multiple point-sources (that is, municipal wastewater discharges). The bases and assumptions of these conservative exposure scenarios are described in detail in the rapid screening approach document (Environment Canada, 2007a).

The industrial releases scenario and the residential releases scenario differ in relative sensitivity for each substance. In most cases, the residential releases scenario was less sensitive than the industrial releases scenario - that is, substances flagged as potentially causing harm by the residential releases scenario were also flagged by the industrial releases scenario. However, in some cases, the reverse was true. Therefore, for each substance, the result of the more conservative of the two scenarios has been used to identify those substances requiring further assessment.

The industrial releases scenario flagged 205 substances as being of potential concern, while the residential releases scenario flagged 106 substances, with 90 substances being flagged by both. In total, 221 substances (or 21% of the 1062 evaluated) were identified by these scenarios as requiring further assessment.

The detailed spreadsheet associated with this report (Environment Canada, 2007b) includes the calculations and results from application of these scenarios to each substance.

RAIDAR

The multimedia fugacity-based RAIDAR (risk assessment, identification and ranking) model accounts for the combined consideration of a substance's physical/chemical and hazard properties and allows evaluation of potential for harm to organisms in different environmental media as well as in food chains. It also accounts for release of a substance over a larger geographic area through its full life-cycle. The approach and its application to a number of substances on the DSL are described in detail in a report produced by the Canadian Environmental Modelling Centre (CEMC, 2007a). The salient features and limitations of RAIDAR in the context of rapid screening are also discussed in Environment Canada 2007a.

As discussed in the above mentioned reports, RAIDAR and similar models are not applicable to all categories of substances encountered on the DSL. Of the 14 chemical categories described in CEMC 2007a, RAIDAR has been applied to substances in five categories - conventional organics; dissociating organic acids; dissociating organic bases; gases; and involatile organics. Therefore, of the 1062 substances evaluated in Step 2, 508 (or 48%) have been modelled using RAIDAR.

As described in Environment Canada 2007a, the Level III model scenario that assumes 33% release of the substance to each of air, water and soil has been selected for interpretation of RAIDAR results in rapid screening. The outputs from RAIDAR of greatest relevance to this work include the media of concern, the critical emission rate (Ec) and the risk assessment factor (RAF). The Ec is the minimum quantity of a substance that would have to be released to a unit world (a region of fixed dimensions) in one year in order for the most sensitive organism in that world to be exposed to a concentration that will cause a harmful effect. The RAF is the ratio of the actual emission rate to the critical emission rate for the substance. In rapid screening, the quantity of the substance believed to be in commerce is conservatively assumed to be the actual emission rate (that is, it is assumed that 100% of the substance is released to the environment over its life-cycle).

A spreadsheet associated with the CEMC report includes all input values and results from application of RAIDAR to these substances (CEMC, 2007b). It should be noted, however, that when the RAF was initially calculated, it was assumed that the quantity in commerce for all substances was 316 kg (the geometric mean of 100 kg and 1000 kg). The detailed results spreadsheet for rapid screening (Environment Canada, 2007b) provides an RAF value for each modelled substance that has been recalculated based on the actual quantity believed to be in commerce in Canada, as was discussed in the section on "Substances covered under this assessment".

As with other models, results from RAIDAR depend on the quality and quantity of the available substance-specific data. The CEMC report discusses some specific issues relating to uncertainties in input data. A separate document (Environment Canada, 2007d) discusses how the results of RAIDAR were interpreted with these considerations.

In order to identify which substances are unlikely to have the potential to cause ecological harm, it is necessary to select a cut-off value for the RAF. A value of 0.001 was selected, which is equivalent to an uncertainty factor of 1000. Selection of this conservative value allows for up to a 1000-fold error in the model results owing to uncertainties in the quantity of the substance in commerce and other model inputs, such as physical-chemical properties. The ability of RAIDAR to discriminate potential for ecological harm based on the characteristics of substances is discussed further in Environment Canada 2007d.

Based on the described model scenario and the selected RAF cut-off value, 26 of the 508 substances that were evaluated using RAIDAR were identified as requiring further assessment. Of these, 21 had also been identified by the generic aquatic scenarios discussed above. Therefore, an additional 5 substances were identified as requiring further assessment.

Step 3 - Mechanical filters and manual process

The different sources of information used as mechanical filters are divided in three categories: 1) exposure - quantities, releases and industrial information; 2) hazardous substances lists or substance profiles; and 3) miscellaneous properties and hazard databases. As described in the rapid screening approach document (Environment Canada, 2007a), each of the information sources has been allocated a weighting that reflects its relevance to the context of this evaluation. Some sources are considered of sufficient relevance that substances that they include are immediately identified under rapid screening as requiring further assessment. Others, that are either considered of lesser relevance or for which more detailed evaluation of substance-specific data contained in the information source is possible, are directed to the manual process stage for a decision to be made based on the weight of the full range of available information.

Appendix A shows the number of substances that have been flagged by each of the mechanical filters for the 836 substances that were evaluated at this stage of rapid screening. As is seen, some mechanical filters have flagged more substances than others. Twenty-nine of the substances appear on one or more of the six international lists of High Production Volume (HPV) chemicals. As this demonstrates that these substances are in use in high quantities in at least some countries, the assumption that they are currently used in very low quantities in Canada is being questioned. As a result, they have been identified as requiring further assessment, which may include collection of current information on quantities in commerce. Exceptionally, 3 substances that are on an HPV list but for which recent Canadian data has already been collected, are further evaluated at the manual process stage. A further 387 substances were flagged by other mechanical filters, leading to their evaluation at the manual process stage. A total of 390 substances were evaluated at the manual process stage.

Substance-by-substance evaluation at the manual process stage is based on consideration of the weight of evidence from available information as to whether the substance may currently be in use in Canada in quantities greater than had been assumed, or if the substance has hazard properties or characteristics that may not have been adequately addressed using the exposure scenarios in step 2. This may involve looking at temporal trends in the quantity of a substance in commerce in other countries.

As a result of this further evaluation, 53 substances were identified as requiring further screening assessment. A summary of the basis for the decision on each of the 390 substances evaluated using the manual process is presented in the detailed results spreadsheet (Environment Canada, 2007b).

Page details

Date modified: