The relationship between site history and human health risks: Lessons from 13 years of contaminated land risk assessments in Victoria, Australia

Site history is widely recognized as one of the key determinants of the nature and scale of contamination that is likely to be present at a location. While there is vast anecdotal and empirical evidence on the relationship between site history and risk of exposure to pollution, few attempts have been made to systematically collate data across sites to rigorously identify such trends. This study presents an analysis of human health risk metrics extracted from all 1732 contaminated land environmental audits published in Victoria, Australia between 2006 and 2018. The contaminating activities found to be most likely associated with elevated risks to human health were service stations, fuel depots, dry cleaners, gasworks, mechanical parts manufacturers, imported fill and unknown offsite sources generating regional groundwater pollution (usually polluted with trichloroethene). The ten chemical contaminants most frequently assessed in human health risk assessment reports included (in order): benzene, C>10–16 petroleum hydrocarbons, naphthalene, C6–10 petroleum hydrocarbons, xylenes, trichloroethene, toluene, ethylbenzene, benzo(a)pyrene, and tetrachloroethene. The findings of this report include recommendations to develop regulatory guideline values in Australia for trimethylbenzenes, heavy fraction petroleum hydrocarbons (C>16), and trans‐1,2‐dicholoethene, and to strengthen the evidence base informing risks associated with light fraction petroleum hydrocarbons (C6–16) and trichloroethene. To the author's knowledge, this study is the first to extract risk metrics from environmental audit reports and provides a strong evidence base to help regulators, academics and commercial practitioners rank and prioritize sites based on their site history and associated risks.


| INTRODUCTION
Historical industrial, agricultural and other uses of land have resulted in the widespread contamination of soil and groundwater on a global scale.In many cases, pollution arising from such activities is persistent and results in legacy issues that can last for decades or even centuries after the contaminating activity has ceased.Contaminated land has been recognized internationally as a major threat to the health of soils and of associated ecosystems and people (FAO and UNEP, 2021).On a more local scale, in Australia, there are an estimated 160,000 contaminated sites (Fergusson, 2017).In early colonial Australian history, the state of Victoria was a key centre of Australian manufacturing (Linge, 1979).For over 100 years from colonization and through the peak of industrialization , there was no or limited environmental regulation relating to industrial waste discharges to the environment or clean-up of former industrial sites at the end of their functional life.This has resulted in widespread legacy contamination in Victorian land, groundwater and other environmental media.
In 1970, the Environment Protection Act Victoria (the EP Act, 1970) was introduced, but it was only in 1990 that it was amended to incorporate a system of environmental auditing targeted at addressing the issue of legacy contamination.The system relies on environmental auditors, who are independent individuals with a minimum of 8 years of relevant experience, appointed in a statutory capacity by the Environment Protection Authority Victoria (EPA Victoria).
Environmental audits are written and published under the EP Act, 1970, which has since been replaced by the EP Amendment Act 2017 which came into effect in 2021.Not all contaminated land in Victoria requires an audit report, and audits are typically requested by EPA Victoria or by local city councils, often when land is being redeveloped to more sensitive land use, or when the land in question is subject to an Environmental Audit Overlay.
Environmental audit reports are publicly available documents that are accessible through the EPA Victoria website.They are usually large reports that are required to include not only the audit report itself but also all the various environmental assessment reports that were considered by the environmental auditor in the preparation of their audit.These are required to be carried out in accordance with the National Environment Protection (Assessment of Site Contamination) Measure 1999 (NEPM ASC) (which was amended in 2013) (NEPC, 2013).Of particular interest to the present research are 'human health risk assessment' (HHRA) reports, often included among the attachments of environmental audit reports.These assessments serve the purpose of objectively and systematically addressing the potential risks to human health posed by the contamination at the site, and in most cases do so in accordance with published methodologies such as those in enHealth (2012).
A key objective of this study was to rank and prioritize the level of risk posed by different types of contaminated sites on human health, and this was done by extracting risk scores from environmental audits over a 13-year period.While other publications have relied on data from publicly available Environmental Audits (e.g.Mikkonen et al., 2016), this is the first to specifically extract and use data from published HHRAs, with the objective of presenting an overview of the key risks to human health posed by contaminated land in Victoria.While a better understanding of the link between site history and the risk could never replace the need for site-specific sampling and assessment, it is intended to help prioritize additional research and policy work on key environmental pollutants and on specific sources of contamination found to be associated with a higher level of risk to human populations.
This study focussed on risks to human health.While risks to ecosystems are potentially just as important, they were not considered here as the data provided in the audits did not support the assessment approach adopted in this research.

| Audit report selection and data collation
All environmental audit reports available on the EPA Victoria website and dated between 2006 and 2018 (inclusive) were included in this assessment, for a total of 1764 reports.Of these, 21 could not be accessed from the EPA online database, and 11 were not searchable and were therefore excluded from consideration.The following information from the remaining 1732 audits was manually inputted into a database: a unique identifier of each audit (referred to as a 'CARMS' number), EPA transaction number, audit address, date, auditor name, and a summary of the site history (whenever possible included verbatim as provided by the auditor).
In addition, these reports were searched for any HHRA section or attachment, defined here as any analysis involving the calculation of a hazard quotient (HQ), Hazard Index (HI) or incremental lifetime cancer risk (ILCR), or any measurement or estimation of contaminant bioavailability or bioaccessibility to humans.This was done by searching each document for the following terms: 'HHRA', 'hazard quotient', 'hazard index', 'cancer risk', 'bioaccess*' and 'bioavail*'.
Of the 1732 audit reports that could be accessed, 376 were found to include an HHRA section or attachment.The following data were manually extracted from these reports: (a) details of each chemical of potential concern (CoPCs) that was the subject of further analysis in the HHRA; (b) whether the audit related to contaminated land or not; (c) whether the contamination was deemed to be originating from the site or from an off-site source; (d) the principal contaminating activity identified in the HHRA, categorized into one of the activities described in the General Environmental Practice Note (DSE, 2021) as appropriate.In situations where the site history included multiple activities, the contents of the audit report and HHRA were used to determine the most likely or most significant contaminating activity.In situations where the contaminating activity was not known, it was labelled as 'unknown'.The selection of CoPCs in HHRA reports is the result of the chemical concentrations in environmental media, the exposure pathways that are expected to occur on a site, the availability of guideline values to provide an initial assessment of the measured concentrations and the professional judgement of the risk assessor carrying out the assessment.Despite the variations that may exist in the approaches used by different risk assessors to select CoPCs for a site, it is reasonable to generally assume that the CoPC data collected in this study is representative of chemical substances that, through a screening process, were deemed to warrant further assessment through an HHRA.
In addition to the information above, data were also collated for each audit report on the highest HQ, HI or ILCR for any of the following exposure scenario types: (a) any residential scenario, (b) intrusive maintenance workers, (c) construction workers, and (d) any other type of occupational exposure (usually in a commercial setting).Risk scores were converted to the following categorical variables: very low (V.LOW) risks were associated with HI <0.1 or ILCR <10 −6 ; 'LOW' with 0.1< HI ≤1 or 10 −6 < ILCR ≤10 −5 ; moderate ('MOD') with 1< HI ≤10 or 10 −5 < ILCR ≤10 −4 ; 'HIGH' with 10< HI ≤100 or 10 −4 < ILCR ≤10 −3 and very high ('V.HIGH') with HI >100 or ILCR>10 −3 .This 5-point scale was used to help describe threshold and non-threshold human health risks on comparable scales.Effectively, very low and low risks are both within acceptable limits (noting that the acceptable cancer risk in Australia is 1 in 100,000-NEPC, 2013).Moderate, high and very high risks are, respectively one, two or three orders of magnitude above acceptable limits.This data matrix was used to identify the temporal and frequency trends in the reliance of HHRAs, the proportions of HHRA reports addressing each CoPC, risk score of contaminating activities and the principal contaminating activities.

| Data simplification for establishing the relationship between contaminating activities and CoPC
Some of the analyses conducted in this study aimed to investigate the relationship between contaminating activities and CoPC, and this required further simplification of the available data matrix.Each of the 376 HHRAs was identified using their unique CARMs number.The CoPC data per HHRA was converted to binary information where 1 indicated that the CoPC had been identified in the HHRA and 0 indicated it had not been.In this way, CoPC were considered regardless of the magnitude of the HQ or ILCR.The contaminating activity for each HHRA was recorded as a grouping variable.To enable computation and reduce the impact of uncommon site histories on the statistical analysis, the matrix was reduced to only include CoPCs identified in at least 10% of all HHRAs.Specifically, the following 22 CoPCs were retained: 1,1-dichloroethene, 1,2,4-trimethylbenzene (1,2,4-TMB), 1,2-dichloroethane, arsenic (As), benzene (B), benzo(a)pyrene (BaP), chloroform, cis-1,2-dichloroethene (cis-1,2-DCE), ethylene (E), lead (Pb), naphthalene (NAP), tetrachloroethene (PCE), toluene (T), total polycyclic aromatic hydrocarbons (PAHs), total recoverable hydrocarbon (TPH) in the C 6-10 , C >10-16 , C >16-34 and C >34-40 fractions, trans-1,2dichloroethene (trans-1,2-DCE), trichloroethene (TCE), vinyl chloride (VC) and xylene (X).
In addition, only audit reports with one of the following contaminating activities were considered (and their abbreviations provided in brackets): automotive repair/engine works (AUTO), chemical manufacturing/storage/blending (CHEM), dry cleaning (DRYC), filling with imported soil (FILL), fuel storage depot (FUEL), gasworks (GASW), mechanical parts manufacturing (MECH), mining and extractive industries (MINE), infilled quarries (QRRY), service stations (SERVO), textile operations (TEXT), underground storage tank not associated with contaminating activities listed above (UST) and sites where the source of the contamination was unknown (UNKN).These site histories were selected as there was a minimum of 8 HHRA reports available for each site history.Ultimately, 288 HHRA reports were used in this analysis.

| Statistical analyses
Statistical analyses were performed using PRIMER 7 (Version 7.0.17)with PERMANOVA +1 add-on (Clarke & Gorley, 2015) multivariate statistics software package.The final 288 HHRA reports were used to construct a Jaccard resemblance matrix of 17 CoPCs which was subject to visualization using a dendrogram created via hierarchical agglomerative clustering with group-average Linking (CLUSTER) and associated Similarity Profile (SIMPROF) analysis.At each node of the dendrogram, an SIMPROF test was performed to assess if the internal structure of the particular group of samples contained any significant internal structure, thereby indicating where further grouping of the data was no longer warranted.The null hypothesis of the SIMPROF test was that there was no significant difference between the CoPCs identified in the HHRAs within a group; this hypothesis was rejected if the significance level associated with the SIMPROF test statistic was <5%.

| Temporal trends in audits and reliance on human health risk assessments
The temporal trends of the 1764 audit reports that were accessible for the years between 2006 and 2018 show an apparent spike in audits not relating to contaminated land in 2014, but this is thought to be because this type of audit only started to be published on the EPA Victoria website in 2014, and any prior non-contaminated land audits are simply not available.There appeared to be only minor yearto-year variations in the total number of contaminated land audits (Figure 1).The only apparent exception to this is that contaminated land audits appeared to peak around 2014, before gradually decreasing back to long-term average levels in 2017.The exact reasons for this are not known, but it is plausible that this may be related to the publication of an amendment to the ASC NEPM (1999) in 2013 (NEPC, 2013).The proportion of contaminated land audits that included an HHRA varied between 20.4% in 2008 and 34.7% in 2014, with the average percentage of such audits at 27.4% over the period between 2006 and 2018.

| Frequency and risk score of contaminating activities
A total of 29 contaminating activities were identified in the course of this research (Figure 2) with a small number of the activities accounting for the majority of all HHRAs.Thus, the top 8 contaminating activities accounted for 81.6% of all the HHRAs that were reviewed.The most common contaminating activities were service stations and the use of imported fill on the site, followed by instances when the contaminating activity was unknown (in most cases associated with groundwater).
The most likely contaminating activities to result in an unacceptable risk score (i.e.moderate or higher) were underground storage tanks (56% probability), mechanical parts manufacturing (44%) and fuel storage depots (40%) (Table 1).While unacceptable risk scores are generally cause for concern, 'very high' risk scores are particularly important as they exceed acceptable risk levels by at least a hundred times.Only 21 of HHRAs identified a 'very high' risk to human health in at least one metric.When expressed for each contaminating activity as a proportion of all HHRAs for that activity, most 'very high' risk scores were associated with fuel storage depots (30%), dry F I G U R E 1 Temporal trends in the audit reports that were reviewed.cleaning (17%), mechanical parts manufacturing (13%), service stations (9%), chemical manufacturing (8%), textile operations (8%) and gasworks (3%).
The most likely contaminating activity to require an HHRA was gasworks (78% of all audits on gasworks included an HHRA) (Table 1), which is consistent with gasworks usually being large, complex sites often requiring a detailed evaluation of risks.Other contaminating activities that frequently included HHRAs in their audit reports included dry cleaning (for which 44.4% of audits included an HHRA), and service station (39.6%).
Most of the common CoPCs associated with each contamination activity are consistent with the types of pollutants that would be expected on the relevant sites, such as hydrocarbons and BTEX at service stations and fuel depots, and PAH-containing tars at gasworks sites (Table 2).

| Chemicals of potential concern
A total of 127 CoPCs were identified in all the HHRAs that were reviewed.A small number of CoPCs were very common in different HHRA reports, with the three most common ones being benzene (in 39.4% of HHRAs), C >10-16 TPHs (in 39.1% of HHRAs) and naphthalene (in 33.2% of HHRAs) (Table 2).Other common CoPCs included monoaromatic hydrocarbons, common polycyclic aromatic hydrocarbons (NAP and B(a)P), light fraction hydrocarbons and chlorinated solvents (TCE and PCE).Also of note is that the majority of CoPCs were only relevant to a small number of HHRAs, with 44 of the CoPCs only being included in a single HHRA report across the entire 13-year assessment period of this study.While the more common CoPCs in most cases had one or more guideline values (Table 2), it is noted that The frequency with which each principal contaminating activity was identified out of 376 human health risk assessment reports.

T A B L E 1
The eleven most frequently identified principal contaminating activities in human health risk assessments (HHRAs), and their commonly associated chemicals of potential concern (CoPCs), the distribution of the risk scores reported in HHRA and the number of HHRAs reporting on these activities over time.T A B L E 1 Continued (Continues) some common CoPCs do not have any health-based screening-level criteria in Australia, and these include C >16-34 TPHs (found in 14.9% of all HHRAs), 1,2,4-trimthylbenzene (10.4% of HHRAs), C >34-40 TPHs (9.3% of HHRAs) and trans-1,2-dichloroethene (5.6% of HHRAs).

| Relationship between site history and chemicals of potential concern
Multivariate statistics were used to develop a dendrogram showing the clustering of sites that had similar CoPCs, based on the 288 HHRAs that were used in this analysis.Clustering was found between contaminant activities and CoPC with eight key clusters identified (Figure 3).Cluster A showed a correlation between arsenic as a CoPC and mining as the contaminating activity, which is likely because of the history of gold mining in many parts of Victoria and the associated arsenic contamination (Smith et al., 2003).Clusters B and C were associated primarily with chlorinated solvents, toluene and ethylbenzene, and related primarily to contaminating activities that involve solvent uses like mechanical parts manufacturing, chemicals manufacturing and dry-cleaning.These clusters also included most of the HHRAs where the source was unknown.Clusters D and E were primarily associated with fill material, and to a lesser extent gasworks, with the key indicator CoPCs being benzene, benzo(a)pyrene, lead and naphthalene.Finally, Clusters F, G and H were primarily associated with hydrocarbon contamination including BTEX, petroleum hydrocarbons and PAHs.The key differences between clusters G and H were related to the molecular weight of the hydrocarbons, with cluster G (mostly gasworks) including a greater proportion of heavier hydrocarbons, while cluster H (mostly service stations) included lighter hydrocarbons such as BTEX, gasoline and diesel.Cluster F was very similar to G and H but was characterized by the cooccurrence of petroleum hydrocarbon contamination and chlorinated solvent contamination, originating either from the types of activities that occurred historically on the sites or as a result of unknown offsite contamination sources.

| Risk profile of contaminating activities
There is no agreed measure of risk that can be universally applied to compare health risks from different sites.The most direct way of rating health risks would be to carry out extensive epidemiological studies on health effects associated with contaminated sites, but these studies T A B L E 1 Continued are notoriously complex and difficult to complete (Hoek et al., 2018).To date, only a small number of these epidemiological studies has been completed, including one that identified significant effects of TCE exposure on birth outcomes in New York (Forand et al., 2012), and one that identified no increase in cancers associated with a gasworks site in Illinois (Alexander et al. 2014).
We adopted three alternative indicators of risk to help rank the 13 contaminating activities that were most frequently the subject of detailed human health risk assessment (Table 1).When considered together, these three approaches can provide a more holistic view of the risks posed by the assessed sites than any indicator would do on its own.The first and simplest approach simply involves ranking based on the absolute number of audits requiring risk assessments for each contaminating activity.This approach assumes that sites requiring an HHRA are more likely to pose a significant risk and that more common types of contaminating activities have a higher likelihood of exposing more people to hazardous contaminants.The contaminating activities that most often required a risk assessment report were service stations, followed by importing contaminated fill and contaminated groundwater reaching the site from an unknown source.
The second approach to ranking the risk posed by different contaminating activities is the relative proportion of audits requiring an HHRA component for each activity.Activities that more frequently require HHRAs are more likely to result in significant contamination, while inherently low-risk sites would not typically require an HHRA except under exceptional circumstances.Based on this second approach, gasworks clearly pose the greatest level of risk, followed by dry cleaners and service stations.
Finally, the third approach to ranking contaminating activities is based on the reported order of magnitude of the risk scores reported in the various HHRAs.The contaminating activities with the highest probability of resulting in an unacceptable risk are underground storage tanks, mechanical parts manufacturing and fuel storage depots.Of these, only a small subset of HHRAs reported 'very high' risk scores, but these warrant further discussion as they are the most likely to result in actual, measurable health impacts in exposed humans.The types of contaminating activities that resulted in such high-risk scores were fuel storage depots, dry cleaners, mechanical parts manufacturing, service stations, chemical manufacturing, textile operations and gasworks.These contaminating activities pose the greatest risk of creating measurable health impacts in exposed humans.
Overall, when the results of the three different ranking approaches are combined, the findings of this paper suggest that the key contaminating activities with a potential for realized public health impacts include service stations, fuel storage depots, dry cleaners, gasworks, mechanical parts manufacturers, imported fill and polluted groundwater originating from one or more 'unknown' sources.In 84% of HHRAs where the contaminating activity was unknown, the CoPC was TCE.This pattern is consistent with the historical widespread use of TCE as an industrial solvent (ATSDR, 2001;Doherty, 2014), the persistence of TCE in oxygen-poor environments like sub-slab soils and deep aquifers (ASTDR, 2001), and its widespread distribution in groundwater (IARC, 2014).While frequent, these unknown sources of contamination did not result in very high risk scores and only resulted in unacceptable risks 13% of sites.This suggests that while TCE contamination can often reach a site from unknown offsite sources, it only poses an unacceptable risk in a small number of locations.While site history can provide a strong indicator of the health risks of a site, our research also shows that unknown offsite sources can be an important consideration.

| Risks associated with unknown sources
A contaminating activity that was of particular interest in this research was the 'unknown' category (i.e. when

Note:
The subscripts S, G and V are used to denote, respectively, CoPCs for which a soil, groundwater or soil vapour human health guideline value is provided in the ASC NEPM (NEPC, 2013).Such criteria include health investigation levels, health screening levels, and groundwater investigation levels (drinking water).
no specific source could be identified).In 84% of HHRAs where the contaminating activity was not known, the CoPC was TCE.As stated above, this pattern is consistent with the historical use of TCE as well as its physical properties.When considering the fact that unknown was the third most common contaminating activity (32 HHRAs) the importance of chlorinated solvents as historical, diffuse substances with no readily identifiable source is apparent.
While frequent, these unknown sources of contamination never resulted in very high-risk scores, and only resulted in unacceptable risks in up in 13% of sites.The dendrogram (Figure 3) showed that the CoPC mixture typical of sites with unknown contamination sources was most consistent with the signature of chemical manufacturers, dry cleaners and mechanical parts manufacturers, suggesting that these are plausible sources of 'unknown contamination'.
The ubiquitous nature of TCE and its potential to impact the health of exposed individuals have been noted by other authors (e.g.ATSDR, 2001;De Miranda & Greenamyre, 2020;IARC, 2014) and provide support to the view that clearer regulation of TCE and other chlorinated solvents in Victoria would be beneficial.Under the current NEPM ASC (NEPC, 2013), assessors are not provided with a full complement of screening guideline values.The findings of this research suggest that chlorinated hydrocarbons often drive HHRAs but infrequently represent unacceptable risks.Therefore, the development of additional guideline values for trichloroethene and other chlorinated hydrocarbons in groundwater, soil vapour and other media would be beneficial.

| Contaminants of potential concern in HHRAs
While a small number of pollutants commonly drove HHRAs, the vast majority of pollutants (83%) were uncommon and only selected in less than 5% of these reports (Table 2).This pattern in CoPCs is unsurprising as it is a function of the frequency of occurrence of pollutants and the potential risks they pose to human health.More volatile substances were frequently identified among the more common CoPCs, and this is thought to be at least in part owing to the typical exposure pathway; significant chronic exposure to volatile substances is more likely than significant direct exposures via soil ingestion or dermal contact.
Another factor that is thought to influence the frequency of CoPCs being adopted in HHRAs is not related to either the prevalence of these substances in the environment or the level of risk they pose to human health.Contaminated land assessments invariably involve an initial evaluation of measured pollutant concentrations against guideline values, such as the health investigation levels (HILs) or health screening levels (HSLs) specified in ASC NEPM (2013).When the HILs or HSLs are particularly conservative or prone to variations owing to site-specific factors, that creates a stronger incentive for the preparation of an HHRA report.It is particularly worth noting that the interim HILs in Australia for chlorinated hydrocarbons are acknowledged as being highly conservative (Ma et al., 2020;NEPC, 2013), so it is not surprising that they feature heavily among the most common CoPCs in HHRA reports.
Together, the factors listed above drive the frequency of the CoPCs shown in this paper.The most frequent CoPCs are substances that are commonly encountered at contaminated sites, are more likely than not to be volatile or otherwise pose a potential risk to human health and may be associated with guideline values that are either over-protective or not particularly representative of the exposures at the site.
The top 10 CoPCs that were most frequently selected in HHRAs were (in order): benzene, C >10-16 TPHs, naphthalene, C 6-10 TPHs, xylenes, trichloroethene, toluene, ethylbenzene, benzo(a)pyrene and tetrachloroethene.While this list applies specifically to contaminated land and groundwater, it can be compared with other similar lists of priority substances.The European Union lists benzene, naphthalene, trichloroethene, tetrachloroethene and benzo When comparison is made between the most common CoPCs identified in this study and the other rankings of chemicals, there is a noticeable gap in the priority that is given by various agencies to better understanding the risks posed by light fraction petroleum hydrocarbons (C 6-10 and C >10-16 ).Although these chemical mixtures may not be inherently as toxic to humans as some other substances, their widespread distribution in the environment and the fact that they are relatively volatile mean that the importance of rigorous, high-quality data to inform decisions around these CoPCs should not be underestimated.In Australia, Friebel and Nadebaum (2011) developed the health-based guideline values for C 6-10 and C >10-16 TPHs and demonstrated the great complexities and multiple assumptions that are involved in the assessment of these substances.It has been suggested that the toxicity of petroleum hydrocarbons is not fully understood and should be the target of more research (Kuppusamy et al., 2020).The results of our study support this view.
If it is reasonable to assume that CoPCs in HHRA reports are to be thought of as substances warranting further investigation, then the frequency distribution of these pollutants can be used to inform environmental policy to help prioritize the development or refinement of guideline values.In particular, some of the most common CoPCs for which no relevant human health-based guideline values are provided include 1,2,4-trimethylbenzene, C >16-34 and C >34-40 petroleum hydrocarbons, and trans-1,2-dichloroethene.Toxicologists, risk assessors, regulators and policywriters may wish to consider the list presented in this paper as a guide to help prioritize pollutants to ensure that toxicology or policy updates would create the greatest impact.
Likewise, pollutants not listed in this report have not been the subject of even a single risk assessment in 13 years.This suggests that unless a substantial change occurs, they are unlikely to warrant significant attention in the future.An example of a recent substantial change would be the recognition over the last approximately 5 years of per-and poly-fluoroalkyl substances (PFAS) as widespread chemicals on concern in Australia.

| Conclusion
This study presents a novel approach to ranking contaminated site risks based on the contaminating activity that generated the pollution on the site.The use of environmental audit data for this type of analysis is novel and provides a strong evidence base to support the study findings.The contaminating activities that pose the greatest risk were found to be service stations, fuel depots, dry cleaners, gasworks, mechanical parts manufacturers, and imported fill.This research also showed that there is a lack of research and policy guidance on the management of 1,2,4-trimethylbenzene, C >16-34 and C >34-40 petroleum hydrocarbons and trans-1,2-dichloroethene.While the findings of these studies are directly applicable to the state of Victoria, Australia, they are likely to be representative of many other locations worldwide, and in particular locations that experienced a large surge in manufacturing industries over the last century.

F
Dendrogram shows the relationships between site activities and chemicals of potential concern in human health risk assessments in Victoria, Australia.
Proportions of HHRA reports addressing each CoPC.
(a)pyrene among the 33 priority substances for environmental quality standards (European Parliament, 2008).The European Chemical Agency (ECHA) only lists one of the top 10 CoPCs among its substances of very high concern (trichloroethene, ECHA, 2008).Likewise, the Agency for Toxic Substances and Disease Registry 2019 Substance Priority List includes benzene as having a rank of 6, benzo(a)pyrene with a rank of 8, and all but two of the top 10 CoPCs identified in this study among the substances with ranks up to 170 (the two exceptions being C 6-10 and C >10-16 TPHs) (ATSDR, 2019).Under the Toxic Substances Control Act, the United States Environmental Protection Agency has developed current risk management activities for only two of the top 10 CoPCs identified in this study, tetrachloroethene and trichloroethene (U.S. EPA, 2021).Under the Canadian Environment Protection Act 1999, the Minister of Environment and Climate Change Canada and the Minister of Health are required to establish a priority substances list (PSL).Two PSLs have been published (one in 1989 and a second one in 1995), and they include all the top 10 CoPCs identified in this study except for petroleum hydrocarbons (C 6-10 and C >10-16 ), naphthalene and ethylbenzene (Government of Canada, 2019).