«Chemical and Downstream Oil Industries Forum Supplement to Guideline – ‘Environmental Risk Tolerability for COMAH Establishments’ Complex Site ...»
Chemical and Downstream Oil Industries Forum
Supplement to Guideline – ‘Environmental Risk
Tolerability for COMAH Establishments’
Complex Site Example
Whilst the CA cannot comment on the accuracy of any site specific
data or assumptions, the worked example provided does demonstrate
an appropriate interpretation and application of the CDOIF guidance,
with a sufficient level of detail to allow the screening process to be
Supplement to Guideline – ‘Environmental Risk Tolerability for COMAH Establishments’ Complex Site Example v0.0 Page 1 of 35 CDOIF CDOIF is a collaborative venture formed to agree strategic areas for joint industry / trade union / regulator action aimed at delivering health, safety and environmental improvements with cross-sector benefits.
Chemical and Downstream Oil Industries Forum Complex Site Case Study Glossary of Terms used in Case Study APIS – Air Pollution Information System BLEVE – Boiling Liquid Expanding Vapour Explosion CDOIF – Chemical and Downstream Oil Industries Forum COMAH – Control of Major Accident Hazards DEM – Digital Elevation Model EHI - Environmental Harm Index EA – Environment Agency EI – Energy Institute GIS – Geographical Information System HAZID – Hazard Identification (Study) HAZOP – Hazard and Operability (Study) IES - Institute for Environment and Sustainability LOPA – Layers of Protection Analysis MAS – Major Accident Scenario MATTE – Major Accident to the Environment NRW – Natural Resources Wales SAC – Special Area of Conservation SEPA – Scottish Environmental Protection Agency SIL – Safety Integrity Level SSSI – Site of Special Scientific Interest TifALARP – Tolerable if As Low As Reasonably Practicable Overview of Approach It is considered appropriate to review how the current CDOIF guidance might be applied to a complex site where there are multiple sources, pathways and receptors which have the potential to combine to enable the generation of a MATTE. Whilst the CDOIF guidance has been used as the basis for the assessment there are some important deviations from the approach which are necessary to make sure that the assessment remains focussed and presents a meaningful and thorough yet concise output in the context of a complex site.
This worked example follows the CDOIF guidance in terms of the degree of assessment required to demonstrate adequate risk controls are in place. Not all major accident scenarios will be assessed to the same extent, rather they are progressed until the frequency associated with the hazard has been reduced to an acceptable level (or to a point that is not a significant contributor to the overall Establishment risk).
The recommended approach of identifying all of the potential pollutant linkages for a complex site which has multiple potential receptors can lead to a lengthy table of results which ultimately provides only limited value in completing either a qualitative or quantitative assessment of risk to the environment. Instead, the CDOIF approach may be worked in reverse by identifying the most significant receptors (based on proximity, magnitude of impact, sensitivity, etc) and then working through the sources to identify which events could plausibly result in an impact at those locations. For example, where there are multiple designations and only subtle differences in the proximity of the sites then the most onerous receptor in terms of impact area/length thresholds should be selected. Consideration of groundwater is a critical element and it may require expertise to assess whether that particular feature should be considered as a receptor within the confines of the site, beyond the site boundary or whether the groundwater simply constitutes a pathway to convey contamination to a different receptor.
The end point of this first stage is therefore to identify receptors (but without looking in detail at the rationale for designation, etc) and associated sources. This results in a manageable set of potential pollutant linkages for assessment.
The sources can then be characterised through review of the available process safety data to identify those major accident scenarios (MAS) which may have already have been documented as having the potential to result in a source of contamination capable of generating a MATTE. This list of sources should then be critically reviewed to remove those unlikely to have a MATTE potential and supplemented by additional scenarios which may not have been considered as part of previous safety assessment work (e.g. tank floor failures).
For each asset there may be a range of plausible sources to the same receptor. To simplify the subsequent calculation steps each asset is assigned to its own compartment. Splitting each part of the site in this way will enable the process of assigning risks to be made transparent and can be effectively managed in a combined spreadsheet and geographic information system (GIS).
Having identified a potential set of sources and receptors the next stage is to identify what types of pathways might join the two – whether that be (for example) via overland flow routes, subsurface migration or via emissions to the atmosphere.
Some pathways may be dismissed relatively quickly by completing a high level review of the significance of a release whilst others will inevitably require more detailed assessment of initiating frequencies for the release and assessment of the effectiveness of the barriers separating the source from the receptor.
The pathway assessment for a complex site may therefore be completed in three stages;
1. High level assessment to evaluate whether the link from the source to receptor via the defined pathways could result in a plausible MATTE (e.g. fire associated with a tank producing combustion products and its effect on a SSSI receptor via the air pathway).
2. Unmitigated risks taking into account the initiating frequency (i.e. is this already so low that the impact at the receptor is unlikely to be significant) and existing control measures which would limit the potential for a release from primary containment (e.g. Layers of Protection Analysis (LOPA) for bulk storage tanks).
3. Mitigated risk assessment considering the likely effectiveness of measures which would limit the potential for the source to reach the receptor (e.g. secondary/tertiary containment, emergency response plans, in-ground migration and effectiveness of pathway interruption measures, etc).
At the end of this stage the assessment is nearly complete since we have defined the sources and receptors, considered initiating frequencies, built in the engineering controls and considered the measures in place which could limit the chance of a significant quantity of contamination reaching the receptor. The potential level of risk can be viewed on an asset by asset basis (i.e. compartments), for each MAS and which may then be combined for each receptor.
The last stage is to assess the significance of the potential impact. This is left until last as the assessment process itself may assist in understanding how large of an area could be affected following a release. It is also possible to consider a conservative impact level at the unmitigated stage and a different (likely lower) level of impact following a more detailed review. The potential significance and the acceptability criteria to be used in the summation of the establishment risk follows the guidance outlined by CDOIF. The process is relatively straight forward based on the actual receptor(s) identified. This process results in the tolerability criteria being defined for the establishment and from there the results of the summed risks for the Establishment can be compared to this criteria. One potential area which should be addressed is where a site covers multiple hydraulic or surface flow catchments which may have different receptors.
In order to assist in demonstrating the approach an example process is outlined below for assessing the MATTE risk at a refinery site in South Wales.
Introduction The case study which follows is based on the need to develop an assessment of the potential environmental risk posed by plausible major accident scenarios at a refinery, building on the existing process safety assessment and is focussed on completing an assessment of MATTE risks under the COMAH regulations taking into account current guidance from CDOIF.
For context the site is located approximately 3km from the coast and is surrounded by numerous small streams which eventually discharge into an ecologically sensitive, and statutorily designated, site (these are therefore considered as the potential receptors and each was assigned the highest level receptor type based on the ultimate receiving water body which is designated as a SAC). The geology at the site is a mixture of mudstones, siltstones and sandstones which are folded and dip steeply towards the north and south indicating the presence of a syncline through the northern part of the site. The shallowest bedrock unit is typically mudstone which is overlain by a veneer of made ground comprising gravelly clay.
Groundwater beneath the site is classified as being within a Secondary A aquifer with the predominant flow being via fractures and fissures. A groundwater high is located in the north of the site and groundwater flow is generally radial from this point resulting in a range of different receptors for site derived contamination within 5 principal catchments across the main site and a further one at the jetty to the south. Figure 1 provides background to the site setting.
Figure 1 – Environmental Setting Receptor Review The first part of the process is to identify whether there is/are source-pathway-receptor pollutant linkages at the site. This was undertaken in a conservative manner and at a high level. For instance, there are numerous small streams located around
the periphery of the site which drain in to a SAC/SSSI (which have marginally different extents). The site is also underlain by groundwater in a Secondary A aquifer and is surrounded, to a large extent, by farmland. A qualitative appraisal of these receptors indicated that the SAC was likely to be the most sensitive and as such was used as the basis for determining the potential for a MATTE to exist based on the potential aerial extent of impact. At this stage no consideration was taken of the length of adjoining water courses (which might remove them from being MATTE receptors) rather it was assumed these formed an integral part of the SAC and they acted as a pathway which itself was not considered to have any mitigation potential at this stage.
Specific details regarding the SAC/SSSI designations are as follows;
SSSI – The designation is based on a combination of geology and ecology depending on location. Various estimates of the area are provided depending on source information although the formal citation estimates the area to be approximately – 2,190 hectares.
SAC – The SAC covers a very large area. Review of various designations for the SAC indicate that the key Estuary Habitat covers a similar extent to the SSSI (albeit it includes the full area of the water body. In addition the intertidal mudflats and to some extent the Atlantic salt meadows also cover a similar area as the SSSI. This specific component of the overall SAC was considered as the sensitive receptor when working out potential areas of contamination.
Overall the SAC/SSSI designation aspect is relatively complex so a conservative approach was adopted and simply assumed that the SAC was the most sensitive potential receptor. Further information on these and other receptors are provided in the full submission and in other relevant environmental reports for the site.
A set of CDOIF tables which outlines the process for receptor selection is provided later in the case study – once the plausible sources for each MAS have been identified.
Groundwater is a more challenging receptor class for the site given the aquifer designation. It was, however, discounted as being a receptor in its own right for several reasons relating primarily to existing groundwater quality, extent of site ownership, the low likelihood of it being exploited in the future, etc. Instead groundwater was considered as a potential pathway with the various surface water features located around the periphery of the site being considered the primary receptor and which were classified based on their links to the designated site located to the south of the site. Groundwater outside the site boundary was considered a potential receptor but of lower sensitivity than the surface water receptors (i.e. if there was a MATTE potential for groundwater there would also be a potential for a MATTE relating to the surface waters and adjoining SAC).
If groundwater was considered to be a receptor then it would be classified as severity level 2 as Level 3 would require 1km 2 to be contaminated. Given the nature of flow in the bedrock the actual breadth of contamination is likely to be limited and therefore plumes in excess of 2-3km would be required to exceed this lower level threshold. In addition, the submitted report contains a wide range of reasons why groundwater on-site should not be considered a receptor in its own right (acknowledging that there is a wide range of regulation in place to capture contamination of groundwater on site were it to occur). Notwithstanding this it is acknowledged that this receptor could be at risk and may be affected by different MAS pathways and have different mitigation and this should be considered carefully for sites where groundwater may be a significant receptor. Groundwater pathways to the same receptors which may be affected by overland transport of contamination following a release have been assessed and mitigation measures applied separately based on the pathway analysis. For example, penetration into the ground and migration as a dissolved phase plume towards surface water carries with it the potential to mitigate the level of impact and this would clearly not be applicable for overland routes. Similarly the affect of secondary and tertiary containment is nullified if the contaminants could penetrate into the ground and migrate within groundwater. These are all aspects which will be covered in more detail as part of the Stage 3 assessment as needs dictate.
For this complex site the consideration of receptors was stopped at this stage without the need to consider in more detail the specific rationales for designation, species at risk, etc. A conservative view was taken in terms of the magnitude of potential impact which would subsequently be reviewed once information on the MAS and subsequent environmental risk assessment had been undertaken. For this refinery site it was assumed that the majority of the MAS had the potential to impact between 25 to 50% of the area, designated population or associated linear features (Major/Severity Level 3 in the CDOIF classification scheme). The purpose of the tolerability review is to ensure that appropriate tolerability thresholds are used to screen the site risks against. If this is done conservatively there should be no need to consider each potential receptor in more detail than necessary.
Regarding the SAC/SSSI receptors; in this case study professional judgement was used to select the worst case combination of severity and duration which resulted in the SAC being selected from which to assess the relevant severity and duration criteria.. It will be necessary to look at receptors again, in more detail, if there is an intolerable risk and/or if certain mitigations would only work for one receptor and not another. In this case study the mitigation measures applied are applicable to both.
The next step for this case study was then to produce an assessment of environmental risk taking into account the information contained within the existing Process Safety Report to satisfy applicable environmental aspects of the COMAH regulations.
It should be noted that the CDOIF guidance advocates completing the first stage assessment work in a qualitative manner comprising two steps;
1. to establish first whether there is a pollutant linkage and if there is, using information on volumes of product stored etc, determine the potential degree of impact that a site might have on the identified receptors; and