OR/18/012 Scenarios

Loveless, S, Lewis, M A, Bloomfield, J P, Terrington, R, Stuart, M E, and Ward, R S. 2018. 3D groundwater vulnerability. British Geological Survey Internal Report, OR/18/012.

In all case studies, a simple configuration of a vertical borehole drilled to the depth of the hydrocarbon source unit was used, with the option for laterals in the case of shale gas and Coal Bed Methane (CBM). It is recognised that, in reality, hydrocarbon activities are much more complex and the activities and geometries of the sub-surface infrastructure should be assessed according to the best information available.

Examples of the use of the methodology for conceptual scenarios with expected high and low risk are presented. Case studies demonstrate how the 3DGWV methodology could be used as part of site-specific risk screenings and to show the range of results which might be expected in areas of England where hydrocarbon source units exist and are included in Appendix 6. They are based on real data for each of these areas, but the case studies are generic and as such a precise site location has not been specified.

High and low risk examples

The highest risk scenario is likely to involve UCG exploitation as this has been identified as the activity with the highest hazard (Specific vulnerability ). This scenario is shown below. However, since UCG is unlikely to be undertaken onshore in the UK in the near future, the hypothetical highest risk scenarios for shale gas (next highest hazard rating, but constrained to >1000 m bgl) and CBM (lower hazard rating than shale gas but no depth constraints) were also tested.

High risk example, UCG

The high risk scenario for UCG is described in Table 7.1, which is based on the conceptual model in Figure 7.1. The scoring for the intrinsic vulnerability and hazard is shown in Table 7.2 and Table 7.3. The intrinsic and specific vulnerability scores and risk group are summarised at the bottom of Table 7.1. Both the limestone aquifer and coal measures are in the high risk group, and the specific vulnerability score is the highest possible (985).

Figure 7.1    High risk scenario for UCG. Red rectangle indicates UCG activities. Underlying rocks are not assessed in this scenario. Not to scale.

Table 7.1    Scenario and intrinsic and specific vulnerability scores and risk group for UCG high risk example.

Hydrocarbon source and extraction method
Coal measures, UCG
AOI
2 km around vertical borehole
Geological setting
A limestone (fractured) aquifer directly overlies coal measures (Figure 7.1)
Potential receptors	Classification
Limestone aquifer	A (principal aquifer <400 m bgl)
Coal measures	B (secondary aquifer <400 m bgl)
Hazard	Score
Release mechanism of hydrocarbon	UCG, permeability enhancement and increase in pressure and temperature (UCG)
Head gradient driving flow	Upward from coal measures to limestone aquifer
Vulnerability
Vertical separation between source and base of receptor	Calculated from the conceptual model, no lateral change
Lateral separation between source and receptor	Calculated from the conceptual model, no change
Mudstones and clays in intervening units between source and receptor	No intervening units
Groundwater flow mechanism in intervening units between source and receptor, including the receptor	Well connected fractures in both the limestone and coal measures
Faults cutting intervening units and receptor	A fault cuts all units and is known to be transmissive to fluids. This fault also results in the horizontal connectivity of the hydrocarbon source unit and the aquifer.
Solution features in intervening units and receptor	Known to be present in the AOI
Anthropogenic features — mines close to site of interest	Known to be present in the AOI
Anthropogenic features — boreholes close to site of interest	Known to be present in the AOI
Potential receptor	Intrinsic vulnerability score	Specific vulnerability score	Risk group	Confidence
Limestone aquifer	98.5	985	High	Medium
Coal Measures	98.5	985	High	Medium

Table 7.2    Hazard factors for UCG high risk example. Hydrocarbon source is shown in red.

Factor	Release mechanism of hydrocarbon (H1)		Head gradient driving flow (H2)		Hazard score	Confidence
Geological unit	Ranking	Confidence	Rating	Confidence
Limestone aquifer	5	high	2	high	10	high
Coal measures			2	high	10	high

Table 7.3    Intrinsic vulnerability factors for UCG high risk example. Hydrocarbon source is shown in red.

Factor	Vertical separation between source and base of receptor		Lateral separation between source and receptor		Mudstones and clays in intervening units between source and receptor
Weighting (w)	1.5		3		3.5
Confidence	high		medium		high
Geological unit
Limestone aquifer	8	12	4	12	5	17.5
Coal measures	8	12	4	12	5	17.5

Factor	Groundwater flow mechanism in intervening units between source and receptor, including the receptor		Faults cutting intervening units and receptor		Solution features in intervening units and receptor		Anthropogenic features — mines close to site of interest		Anthropogenic features — boreholes close to site of interest		Intrinsic vulnerability score
Weighting (w)	3		4.5		2		8		4
Confidence	high		high		medium		high		high
Geological unit
Limestone aquifer	3	9	4	18	3	6	2	16	2	8	98.5
Coal measures	3	9	4	18	3	6	2	16	2	8	98.5

High risk example, CBM

The high risk scenario for CBM is described in Table 7.4, which is based on the conceptual model in Figure 7.2. The geological setting is the same as for the UCG scenario. The scoring for the hazard is shown in Table 7.5 and intrinsic vulnerability in Table 7.7. The intrinsic and specific vulnerability scores are summarised at the bottom of Table 7.1 . The lower hazard score, due to the release mechanism, results in a lower specific vulnerability score and risk group for both geological units. The limestone aquifer is in the medium/high risk group, and the Coal Measures are in the medium/low risk group. The specific vulnerability scores are 394 for both geological units.

Figure 7.2    High risk scenario for CBM. Thick blue lines indicate CBM wells. Underlying rocks are not significant. Not to scale.

Table 7.4    Scenario and intrinsic and specific vulnerability scores and risk group for CBM high risk example.

Hydrocarbon source and extraction method
Coal Measures, CBM
AOI
2 km around vertical borehole
Geological setting
A limestone (fractured) aquifer directly overlies Coal Measures (Figure 7.2)
Potential receptors	Classification
Limestone aquifer	A (principal aquifer <400 m bgl)
Coal measures	B (secondary aquifer <400 m bgl)
Hazard	Score
Release mechanism of hydrocarbon	Water table lowering and depressurisation (CBM)
Head gradient driving flow	Upward from Coal Measures shale towards limestone aquifer
Vulnerability
Vertical separation between source and base of receptor	Calculated from the conceptual model, no lateral change
Lateral separation between source and receptor	Calculated from the conceptual model, no change
Mudstones and clays in intervening units between source and receptor	No intervening units
Groundwater flow mechanism in intervening units between source and receptor, including the receptor	Well connected fractures in both the limestone and Coal Measures
Faults cutting intervening units and receptor	A nearby fault is transmissive, which results in the horizontal connectivity of the hydrocarbon source unit and the aquifer.
Solution features in intervening units and receptor	Known to be present in the AOI
Anthropogenic features — mines close to site of interest	Known to be present in the AOI
Anthropogenic features — boreholes close to site of interest	Known to be present in the AOI
Potential receptor	Intrinsic vulnerability score	Specific vulnerability score	Risk group	Confidence
Limestone aquifer	98.5	394	Medium/high	Medium
Coal Measures	98.5	394	Medium/low	Medium

Table 7.5    Hazard factor for CBM high risk example. Hydrocarbon source is shown in red.

Factor	Release mechanism of hydrocarbon (H1)		Head gradient driving flow (H2)		Hazard score	Confidence
Geological unit	Ranking	Confidence	Rating	Confidence
Limestone aquifer	2	high	2	high	4	high
Coal measures			2	high	4	high

Table 7.6    Intrinsic vulnerability factors for CBM high risk example. Hydrocarbon source is shown in red.

Factor	Vertical separation between source and base of receptor		Lateral separation between source and receptor		Mudstones and clays in intervening units between source and receptor
Weighting (w)	1.5		3		3.5
Confidence	high		medium		high
Geological unit
Limestone aquifer	8	12	4	12	5	17.5
Coal measures	8	12	4	12	5	17.5

Factor	Groundwater flow mechanism in intervening units between source and receptor, including the receptor		Faults cutting intervening units and receptor		Solution features in intervening units and receptor		Anthropogenic features — mines close to site of interest		Anthropogenic features — boreholes close to site of interest		Intrinsic vulnerability score
Weighting (w)	3		4.5		2		8		4
Confidence	high		high		medium		high		high		high
Geological unit
Limestone aquifer	3	9	4	18	3	6	2	16	2	8	98.5
Coal measures	3	9	4	18	3	6	2	16	2	8	98.5

High risk example, shale gas

The high risk scenario for shale gas is described in Table 7.7, which is based on the conceptual model in Figure 7.3. The scoring for the hazard is shown in Table 7.8 and intrinsic vulnerability in Table 7.9. The intrinsic and specific vulnerability scores are summarised at the bottom of Table 7.7. The intrinsic vulnerability of the limestone aquifer and shale is 98.5 and the specific vulnerability score is 788. Both units are in the high risk group.

Table 7.7    Scenario and intrinsic and specific vulnerability scores and risk group for shale gas high risk example.

Hydrocarbon source and extraction method
Shale, shale gas with high volume hydraulic fracturing
AOI
4 km, 2 km around 2 km lateral boreholes
Geological setting
The minimum depth that high volume hydraulic fracturing is permitted onshore in the UK (1000 m bgl) was used for the depth of the hydrocarbon source unit (shale). Similar to the UCG, a limestone aquifer is the main receptor and directly overlies the shale (the source) (Figure 7.3).
Potential receptors	Classification
Limestone aquifer	A (principal aquifer <400 m bgl)
Coal measures	B (secondary aquifer <400 m bgl)
Hazard	Score
Release mechanism of hydrocarbon (H1)	High volume hydraulic fracturing
Head gradient driving flow (H2)	Upward from shale to limestone aquifer
Vulnerability
Vertical separation between source and base of receptor	Calculated from the conceptual model, no lateral change
Lateral separation between source and receptor	Calculated from the conceptual model, no change
Mudstones and clays in intervening units between source and receptor	No intervening units
Groundwater flow mechanism in intervening units between source and receptor, including the receptor	Well connected fractures in both the limestone and shale
Faults cutting intervening units and receptor	A fault cuts all units and is known to be transmissive to fluids. This fault also results in the horizontal connectivity of the hydrocarbon source unit and the aquifer.
Solution features in intervening units and receptor	Known to be present in the AOI
Anthropogenic features — mines close to site of interest	Known to be present in the AOI
Anthropogenic features — boreholes close to site of interest	Known to be present in the AOI
Potential receptor	Intrinsic vulnerability score	Specific vulnerability score	Risk group	Confidence
Limestone aquifer	98.5	788	High	Medium
Coal Measures	98.5	788	High	Medium

Figure 7.3    High risk scenario for shale gas. Shale unit is pink, thick blue lines indicate planned boreholes. Not to scale.

Table 7.8    Hazard factor for shale gas high risk example. Hydrocarbon source is shown in red.

Factor	Release mechanism of hydrocarbon (H1)		Head gradient driving flow (H2)		Hazard score	Confidence
Geological unit	Ranking	Confidence	Rating	Confidence
Limestone aquifer	4	high	2	high	8	high
Coal measures			2	high	8	high

Table 7.9    Intrinsic vulnerability factors for shale gas high risk example. Hydrocarbon source is shown in red.

Factor	Vertical separation between source and base of receptor		Lateral separation between source and receptor		Mudstones and clays in intervening units between source and receptor
Weighting (w)	1.5		3		3.5
Confidence	high		medium		high
Geological unit
Limestone aquifer	8	12	4	12	5	17.5
Coal measures	8	12	4	12	5	17.5

Factor	Groundwater flow mechanism in intervening units between source and receptor, including the receptor		Faults cutting intervening units and receptor		Solution features in intervening units and receptor		Anthropogenic features — mines close to site of interest		Anthropogenic features — boreholes close to site of interest		Intrinsic vulnerability score (V)
Weighting (w)	3		4.5		2		8		4
Confidence	high		high		medium		high		high		high
Geological unit
Limestone aquifer	3	9	4	18	3	6	2	16	2	8	98.5
Coal measures	3	9	4	18	3	6	2	16	2	8	98.5

Low risk example, conventional oil and gas

The low risk scenario is constructed for conventional oil and gas exploitation since this has been identified as the activity with the lowest hazard. The low risk scenario for conventional oil and gas is described in Table 7.10, which is based on the conceptual model in Figure 7.4. The scoring for the hazard is shown in Table 7.11 and intrinsic vulnerability in Table 7.12. The intrinsic and specific vulnerability scores are summarised at the bottom of Table 7.10. The vulnerability of the sandstone aquifer is 8 and the specific vulnerability score is also 8. A vulnerability of 0 is not possible because the maximum separation distance (>1200 m) and the maximum intervening mudstone thickness (>250 m) have minimum scores of 1.5 and 3.5 respectively.

The risk group is medium/low for this and low for both the mudstone and reservoir. The medium/low risk group, despite a very low specific vulnerability score, reflects the fact that there is a degree of risk to potential receptors with hydrocarbon activities in the subsurface. If the potential receptor was classified as B or C in this case, the risk group would be low.

Table 7.10    Scenario and intrinsic and specific vulnerability scores and risk group
for conventional oil and gas, low risk example.

Hydrocarbon source and extraction method
Conventional oil and gas reservoir, no changes to permeability or pressure
AOI
2 km around vertical borehole
Geological setting
In the low risk scenario a sandstone aquifer overlies 1200 m of mudstones below which overlies a conventional oil and gas reservoir (the hydrocarbon source unit). The aquifer outcrops at the surface (Figure 7.4).
Potential receptors	Classification
Sandstone aquifer	A – principal aquifer <400 m bgl
Mudstone	D – unproductive
Reservoir	C – secondary aquifer >400 m bgl
Hazard	Score
Release mechanism of hydrocarbon (H1)	No permeability enhancement (passive) for conventional oil and gas.
Head gradient driving flow (H2)	No head gradient from source to receptor
Vulnerability
Vertical separation between source and base of receptor	Calculated from the conceptual model, no lateral change
Lateral separation between source and receptor	Calculated from the conceptual model, no change
Mudstones and clays in intervening units between source and receptor	1200 m mudstone in intervening unit
Groundwater flow mechanism in intervening units between source and receptor, including the receptor	Sandstone aquifer and reservoir intergranular flow. No receptors class A to C in intervening units. Therefore, >50% principal or secondary aquifers (EA designation) with intergranular flow.
Faults cutting intervening units and receptor	A fault cuts all units and is known to be transmissive to fluids. This fault also results in the horizontal connectivity of the hydrocarbon source unit and the aquifer.
Solution features in intervening units and receptor	No known solution and no potential for solution features
Anthropogenic features — mines close to site of interest	No mines in AOI
Anthropogenic features — boreholes close to site of interest	No boreholes in AOI
Potential receptor	Intrinsic vulnerability score	Specific vulnerability score	Risk group	Confidence
Sandstone aquifer	8	8	Medium/low	Medium
Mudstone	17	17	Low	Medium
Reservoir	30.5	30.5	Low	Medium

Figure 7.4    Low risk scenario for conventional hydrocarbon activities. Grey is the hydrocarbon reservoir and green is the overlying cap-rock (mudstone). Thick blue lines indicate boreholes. Not to scale.

Table 7.11    Hazard factor for CBM high risk example. Hydrocarbon source is shown in red.

Factor	Release mechanism of hydrocarbon (H1)		Head gradient driving flow (H2)		Hazard score	Confidence
Geological unit	Ranking	Confidence	Rating	Confidence
Sandstone aquifer			1	high	1	high
Mudstone	1	high	1	high	1	high
Reservoir			1	high	1	high

Table 7.12    Intrinsic vulnerability factors for conventional oil and gas low risk example. Hydrocarbon source is shown in red.

Factor	Vertical separation between source and base of receptor		Lateral separation between source and receptor		Mudstones and clays in intervening units between source and receptor
Weighting (w)	1.5		3		3.5
Confidence	high		medium		high
Geological unit
Sandstone aquifer	1	1.5	0	0	1	3.5
Mudstone	1	1.5	4	12	1	3.5
Reservoir	8	12	4	12	1	3.5

Factor	Groundwater flow mechanism in intervening units between source and receptor, including the receptor		Faults cutting intervening units and receptor		Solution features in intervening units and receptor		Anthropogenic features — mines close to site of interest		Anthropogenic features — boreholes close to site of interest		Intrinsic vulnerability score (V)
Weighting (w)	3		4.5		2		8		4
Confidence	high		high		medium		high		high		high
Geological unit
Sandstone aquifer	1	3	0	0	0	0	0	0	0	0	8
Mudstone	0	0	0	0	0	0	0	0	0	0	17
Reservoir	1	3	0	0	0	0	0	0	0	0	30.5

Discussion of methodology

The following discussion is based on the scenarios in Scenarios, and case studies in Appendix 6 – Case studies.

The 3D Groundwater Vulnerability project has developed a prototype Tier 1 (Gormley et al., 2011) methodology for screening vulnerability and risk of groundwater to sub-surface hydrocarbon activities. Screening is site-specific rather than applied across larger areas such as previous EA vulnerability mapping (EA 2017a), due to the large number of inputs and considerations at each site and the variable availability of input data. It provides an indication of the relative risks of hydrocarbon activities in the subsurface to groundwater. The vulnerability and risk parameters and risk group boundaries are, at this stage, preliminary and used for illustrative purposes and it is anticipated that the methodology would be reviewed through experience.

The methodology can be applied as a quick, initial look at possible vulnerability and risk scenarios for a particular development (e.g. assessment of receptors at geological group scale) or as a much more detailed assessment (e.g. assessment of receptors at the geological formation scale). The time taken to undertake a screening using the methodology therefore varies according to the detail required and the purpose of the screening, but also the amount of information available, from one day up to a week or more. Confidence in the classification and groupings improves as more information is brought into the assessment.

The case studies (see Appendix 6 – Case studies) demonstrate how the methodology could be applied in a site-specific setting, including the information that is required and how potential issues may be highlighted. It is not recommended that such initial site-specific risk assessments are decision making tools for regulators, but they could be used to help guide further investigations.

The high and low vulnerability and risk scenarios demonstrate that different geological situations could produce very different intrinsic and specific vulnerability scores. The intrinsic vulnerability is very dependent upon the geometry of the source-pathway-potential receptor system. The specific vulnerability is dependent on the intrinsic vulnerability and the hazard — i.e. the nature of the hydrocarbon activity that is taking place, and the possibility for groundwater flow from the source to the potential receptor. The risk group varies from low to high depending on the specific vulnerability score and the potential receptor value classification. For potential receptors of high value (classified ‘A’), the risk group will always be at least medium/low, recognising that risk can be mitigated but not eliminated when conducting sub-surface hydrocarbon activities. However, this is primarily related to drilling through the formations and not necessarily related to the actual sub-surface activity or the 3D geometry of the system.

The methodology was developed in order to compare the risks posed by conventional, CBM, shale gas and UCG hydrocarbon activities. As such, the vulnerability scores and risk groups are relative. In the current scoring system it is only possible to obtain the maximum specific vulnerability score when conducting UCG activities. However, such activities are unlikely to occur onshore in England in the near future. The high and low risk scenarios show that the maximum score for CBM and shale gas activities is 788 — in the ‘high’ risk group. The highest possible specific vulnerability score for conventional hydrocarbon activities is 197. In this case the highest risk group classification would be ‘medium/low’.

The case studies have shown that accurate potential receptor classification is very important in order to identify a realistic risk group. While classifications based on EA aquifer designations (at outcrop) are reasonable, there can be variations at site-specific scales. For example, the case study for CBM in the West Midlands shows that local information on groundwater quality can be used to downgrade potential receptors (in this case the Sherwood Sandstone and Appleby Group), resulting in a low rather than a medium/low risk grouping for these potential receptors. Similar potential receptor intrinsic and specific vulnerability for CBM exist in the East Midlands. However, the potential receptors remain classified as ‘A’ and therefore the risk group is higher. The case study from Northeast England (Vale of Pickering) indicates that, in some cases, the potential receptors should be upgraded. For example, the Kimmeridge Clay is designated unproductive by the EA, but because it can provide reasonable quantities of potable water it is upgraded to ‘B’. These case studies demonstrate that it is important to compile as much data as possible on both the quantity and the quality of groundwater otherwise misrepresentation may result in overlooked groundwater resources or an overly-conservative view of the risk.

The risk group is the most informative category since it takes into account the sensitivity of the potential receptor. However, intrinsic and specific vulnerability scores must also be consulted to understand the risk group. In all of the case studies, there are a combination of risk groups for the potential receptors in any particular area. Most potential receptors were in the low risk group with the occasional potential receptor in the medium/low risk group. Units in the medium/high risk groups occur rarely, but include principal aquifers overlying shale gas and CBM activities. There are no potential receptors in the high risk group in the case studies under the current classifications (which may be subject to change). Many cases indicated that more information was required to reduce the high levels of uncertainty associated with the risk assessments. In some areas, additional data and information may exist to improve the site specific risk assessments. This should always be taken into account. For example, the information in the Vale of Pickering Methane Baseline Survey (Smedley et al., 2017) could point to natural hydrocarbon migration pathways which would need to be accounted for in the risk assessment methodology.