Graham Cannon makes some useful comments regarding the assessment of carbon dioxide gas in the ground (Talking point, GE July 2002).
It is unwise, however, to treat it as less serious than methane or other gases.
As he rightly points out, there are many locations where low levels of carbon dioxide, for example that generated by bacteria in topsoil, pose negligible risk to developments.
There are others where it can be a significant risk, for example where the source of the gas is abandoned coal workings.
A report by the Department of the Environment in 1995 identified that 25% of mine gas incidents (where mine gas was unexpectedly found at the surface) involved 'blackdamp', a mixture of carbon dioxide and oxygen deficient air.
This can be emitted from the ground at very high rates, particularly in response to rapid and sustained falls in atmospheric pressure, with flows much larger than the volumes exhaled when breathing.
There is one recorded fatality in Widdrington, Northumberland due to seepage of blackdamp from a mine entry into a factory building.
The danger posed by blackdamp was highlighted more recently - a construction worker was killed in 1999 when gas entered a trench near Barnsley, South Yorkshire (New Civil Engineer 3 June 1999).
There are other recorded incidents where it has entered buildings and caused physiological effects, difficulty lighting gas appliances and extinguished pilot lights.
It is therefore important that every ground gas assessment includes a comprehensive desk study to identify likely gas sources and their generation potential.
At this stage many of the small extensions Cannon refers to, where there is no evidence of gas, should be eliminated from the need for further investigation or gas protection measures.
Where more significant sources are identified, a comprehensive ground investigation with gas monitoring will be required to form a conceptual model of gas generation and migration pathways.
Wherever there is a potentially significant source, regular monitoring over three months should be the absolute minimum requirement to identify likely variations in the ground gas regime in the simplest cases.
In complex situations more will be needed. Assessment of the site should then consider all the data together, not just the gas monitoring.
I also agree that risk assessment does not imply complete risk elimination. An important concept in risk assessment, however, is the difference between voluntary risk (for example smoking, or driving) and involuntary risk, such as that posed by gas in the ground.
Society will in general accept much higher levels of voluntary risk and any risk assessment should recognise this.
It is also often equally important to address perceived risk as much as actual risk, particularly where housing is concerned.
Steve Wilson, technical director, the Environmental Protection Group