Prevalence of CO2 from disused mineral mines and the implications for residential buildings: research

5 Previous Incidents

5.1 Incident Record

One of the key objectives of the research project is to collate a record of all the past incidents of CO2 ‘mine gas’ in Scotland that have occurred in the last 54 years which is the time elapsed since national building regulations have come into force.

Annex 1 - Table of past incidents is a summary of all the incidents uncovered in the research project. They have been split into four main categories:

1. Previous incidents of CO2/low O2 in Scotland

2. Other incidents of CO2/low O2 in the rest of the UK

3. CO2/low O2 events elsewhere in the world

4. Other mine gas events in Scotland (where the major component was not identified or believed to be in CO2 or an O2 deficient atmosphere).

All the previous incidents are listed in chronological order starting with the oldest event. Where information is sketchy, or anecdotal, they have been added to the end of the section.

The sources of the case studies came from:

• The Gorebridge IMT report (main section and Appendix 4)
• The Coal Authority (Incidents database)

• Chartered Institute of Environmental Health, The local authority guide to ground gas, 2008

• Local Authorities
• Relevant experts consulted for this report.

Due to the nature of CO2 related incidents, noted earlier in this report, it is mainly the acute incidents, e.g. sudden fatalities or collapse that are reported, and some earlier reports are more anecdotal. Where possible, detail has been given on the incident or mitigation measures since put in place. See Annex 1 - Table of past incidents.

5.2 Observations and Analysis of Previous Incidents

During the collation of previous incidents, similar to that of the Gorebridge event, a number of observations were made from an analysis of the data. Although the data is fairly sparse with no obvious trends, there are some general themes and characteristics which are detailed in the subsections below.

Pathways

On examination of the data, it appears that many of the CO2 ingress events were related to the proximity to old or undiscovered shafts (and adits). This would indicate that old shafts, drifts, and adits are significant pathways for CO2 (and mine gas in general). On occasion this includes unrecorded historic workings such as bell pits. Some of the incidents were related to known shafts which were poorly infilled, indicating that even if there is a known risk, there should not be an assumption that the shaft was sufficiently grouted and there is no gas risk. Several reasons, such as poor initial grouting, further collapse on the edge of the shaft/cavity or general degradation of the original grout may generate a new pathway in what might be regarded as safe, sealed former workings.

Construction

As discussed above it is the proximity of shafts/adits which was an important factor in previous incidents. Within the available datasets there is limited description of the age of the domestic property impacted nor information on construction within the immediate vicinity.

While it has been proposed that previous or ongoing construction activity may increase the risk by altering existing pathways or creating new ones, the data available from previous events is insufficient to assess this.

Data & Reporting

It should be noted that the CA records only date back to 1994 which is when the CA was formed, and a formal database of gas incidents was set-up. In addition to this the information generated by the CA for this report had several errors, such as an incident being listed as non-residential yet in the detail column mentions of effects to ‘residents’. Detail was also lacking on the ‘outcome’ that was recorded. Occasionally the outcome was a very detailed account, but it was mainly limited to basic effects.

Similarly, some incidents are anecdotal and lacking data such as date, location or the nature of the event and therefore also require further verification. In addition to this, based on the nature of CO2 events, it is possible that many chronic events have gone unrecorded, as addressed in the introduction.

Nature of Events

As mentioned previously, CO2 and O2 deficient atmosphere events can be split into two categories, acute and chronic:

• Acute events are typically a fatality or sudden collapse from asphyxiation. Many of these occur when persons enter a subfloor level room or enclosed spaces below ground where significant accumulation of CO2 has occurred.
• Chronic events are indicated by people falling asleep/unconscious, feeling dizzy, or a failure to light a fire or boiler. Flames extinguished in a household therefore indicate a low O2 environment.

Figure 5‑1 CO2/O2 deficient health effects: chronic & acute

Most previous incident reports provided no CO2 or O2 data. In the small subset of domestic properties where this was provided, the information often indicates that an acute event is being investigated e.g. O2 as low as 12% within a domestic property, with one basement area recording 10%.

To give these values some perspective, health effects for depleted O2 levels (Air Products, 2014) occur as follows:

• 19%: Some adverse physiological effects occur, but they may not be noticeable
• 15%–19%: Impaired thinking and attention, increased pulse and breathing rate, reduced coordination, decreased ability to work strenuously, and reduced physical and intellectual performance without awareness
• 12%–15%: Poor judgment, faulty coordination, abnormal fatigue upon exertion, and emotional upset
• 10%–12%: Very poor judgment and coordination, impaired respiration that may cause permanent heart damage, possibility of fainting within a few minutes without warning, and nausea and vomiting
• <10%: Inability to move, fainting almost immediate, loss of consciousness, convulsions, and death.

Of those domestic properties where CO2 levels were available the highest found within a domestic property was 7% or 70,000 ppm. The lowest value noted in the previous incident reports was 0.8% or 8000 ppm CO2.

Again, to provide some perspective health effects for CO2 levels occur as follows:

• 250-400 ppm: Background (normal) outdoor air level
• 350-1,000 ppm: Typical level found in occupied spaces with good air exchange
• 1,000-2,000 ppm: Level associated with complaints of drowsiness and poor air quality
• 2,000-5,000 ppm: Level associated with headaches, sleepiness, and stagnant, stale, stuffy air. Poor concentration, loss of attention, increased heart rate and slight nausea may also be present.
• >5,000 ppm: This indicates unusual air conditions where high levels of other gases could also be present. Toxicity or oxygen deprivation also likely to occur.

From the available data there has been no detailed investigation initiated in domestic properties where CO2 levels were below 5000 ppm. Whilst adjoining properties were subsequently found to have similar issues the reports do not provide data on the gas levels found in these properties. There are few recorded acute events and there is insufficient data to estimate the incidence of lower level chronic effects of CO2 or O2 deficiency within domestic properties (Bonino, 2016).

The symptoms attributed to elevated CO2 or O2 deficiency within a confined space can also be found with many other health issues ranging, for example, from heart problems through to a simple cold. It is, therefore, possible that a low level chronic or transient low-level CO2 event would not be identified as the root cause from a member of the public reporting these symptoms to their doctor.

All previous events recorded at domestic properties have been identified where symptoms are acute or in conjunction with reports of multiple people reporting illness or breathlessness in close proximity.

Mitigation Measures

Where there are mitigation measures recorded against previous occurrences it should be noted that these are a response to an acute event. In most cases the source and pathway were in close proximity to the receptor (i.e. properties) and the reports have been investigated by the Coal Authority with any follow-on mitigation carried out at their direction.

The mitigation measures utilised vary from demolition to passive venting and monitoring, depending on the seriousness of these occurrences. Works completed have included:

• Installation of active venting on nearby shafts
• Continual monitoring
• Existing shafts and adits sealed alongside installation of additional passive venting
• New shafts sunk to access mine workings to allow ventilation
• New boreholes sunk around properties to provide a pathway for ventilation
• Demolition of properties
• Active or passive underfloor ventilation fitted in conjunction with gas membrane
• Shallow workings accessed, and ventilation stacks installed
• Monitors supplied to homeowners.

Where active measures were installed, it is picked up within the Coal Authority’s ongoing monitoring and maintenance regime. It was noted that, within the 1980 occurrence at Cramlington, efforts to mitigate issues at receptor properties using a gas membrane alongside underfloor ventilation were unsuccessful. From the previous occurrences the preference appears to be ventilation in most cases with demolition required where the pathways lead directly underneath the properties.

5.3 Context for Scotland

A key area of the research for this project was to indicate previous incidents in Scotland. A total of 12 incidents were found including the Gorebridge event. Figure 5-2 Distribution of past CO2 events in Scotland shows a map of the distribution around Scotland. Although there are not many recorded incidents in Scotland, CO2 related events have occurred throughout the rest of the UK’s former coalfield areas (detailed in Annex 1 - Table of past incidents).

The CO2 events occurred in South Lanarkshire, Dumfries and Galloway, Midlothian and Fife. From the data, there are three non-CO2 related mine gas events in Scotland. A methane event in Chryston, North Lanarkshire and two incidents in Leven, Fife which were both related to methane accumulation and burning coal.

Figure 5‑2 Distribution of past CO2 events in Scotland