Sir,

Richard Thomas’ letter published in last month’s edition of the AGS Newsletter opens up an interesting debate on how we quantify risk within our profession, especially those subsurface risks that are not readily identifiable or quantifiable. As with any potential risk, we have to address a number of questions, i.e. 1) is the risk real? 2) what are the consequences of the risk? and 3) can the risk be avoided?

Is there a real risk of encountering an Unexploded Bomb (UXB) in areas subjected to WW2 bombing raids? The answer is, despite all the implications, yes. In 1996, the then Armed Forces Minister, Nicholas Soames, released a list of 88 UXBs in the London area of which the Ministry of Defence had records. The location of many of these UXBs were not accurately recorded. In addition, there were many UXBs which went unrecorded altogether during and after the bombing raids of WW2, hence the need to carry out location specific threat assessments looking at all possible sources of information, including anecdotal. It should also be remembered that on many occasions the Luftwaffe aborted their bombing raids on prime targets resulting in indiscriminate bombing as bombs were released to conserve fuel for the journey home.

If the risk is real, how do we quantify that risk. Richard Thomas does not indicate what method he would adopt to differentiate between low, medium and high risk areas for an inherently random and erratic event. Even “low risk” means that there is some potential risk. One therefore has to assess the consequences.

WW2 bombs were designed to cause death and destruction. In recent years proof that time does not diminish a UXBs deadly has been demonstrated in Berlin, Austria and on several construction sites in Japan. A recent tragedy only reinforces the lethal nature of aged Unexploded Ordnance (UXO), only a few months ago 3 Dutch fisherman working near the proposed London Array wind farm site in the Thames Estuary were killed and 2 were badly injured when they hauled up a small “hand grenade like” device which exploded with tragic consequences.

If we were relatives of a victim of an explosion on a construction site, would we be comforted by the fact that the site had been the subject of a probabilistic evaluation and was considered as being “low risk”. I think not. Similarly, the victims of the recent tsunami can draw little or no comfort from the fact that the seabed eruption which resulted in massive destruction was a 1 in a 1000 year event. The “low risk” of occurrence did nothing to diminish the “high risk” consequences of that occurrence.

Traditionally in geotechnics we attempt to minimise risk during the design phase, i.e. engineering out the risk. Such an approach could be adopted in the case of UXBs, i.e. the use of shallow foundations and / or utilisation of any existing foundations. This would, in most cases, be an expensive and impractical alternative. In view of the consequences of the UXB risk both during and after construction and the associated costs of designing out the risk, a well planned threat assessment and survey should be a cost effective option. As a consequence, Fugro Engineering Services Limited have developed their own cone magnetometer technology to provide site specific data upon which the UXB risk can be directly quantified.

What should concern drilling contractors and consultants alike, is the absence of useful regulatory guidance on when and where UXO risk mitigation is needed. The Dutch Government will be introducing legislation regarding UXO surveys later this year. Regulatory guidance will provide the basis for a rational approach to the UXO surveys, focussing attention on risk and consequences rather than the more emotive and subjective topics of cost and probability.

M R Horsnell Director Fugro Engineering Services Limited