UXO Myths within the Marine Environment

It is often a misconception that UXO movement is equal or similar to sediment migration, i.e. is caused by it. The likelihood of an item of UXO migrating along the seabed due to water flow (tidal stream/current) is a function, among others, of seabed composition, firmness and morphology (slopes, ripples, troughs, boulders etc.); the current strength, duration and persistence of direction; and the weight, shape (particularly of protrusions, such as lifting lugs) and orientation of the UXO.

Some smooth, cylindrical types of UXO, such as ground mines and torpedo warheads, have been known to roll along the seabed when conditions are favourable; if the seabed is flat and without obstruction, if it is firm and if the current is strong enough and predominantly uni-directional. If the UXO is laid in shallow water, storm events and tidal surges can also produce the conditions necessary to move UXO from its original position. However, in dynamic sediment conditions where complete or partial mine burial has occurred very soon after lay, which is frequently the case, it is very rare that these conditions are met.

It is very common for fishing trawlers to encounter UXO, either knowingly by bringing it into the vessel in their nets or inadvertently by dragging an item for a distance along the seabed before it eventually falls free. In fact, 50% of finds reported to the OSPAR commission have been due to fishing. Anecdotally, fishermen that have recovered UXO in their nets have also been known to occasionally dump it back into the sea rather than report the incident.

Of note, in reality it is very difficult to quantify this migration mechanism within a risk assessment; mainly because finds are rarely recorded. Those that are, are not usually done so collectively as a coherent archive. The number of encounters and post-find disposal areas cannot therefore be measured with any accuracy. Moreover, unseen, inadvertent movement of UXO – i.e. dragged by a trawl for a distance and then released – is by its nature unquantifiable. Nonetheless, it is important to consider this migration factor as part of the baseline residual risk.

Modern demersal trawls do not penetrate the seabed to any great degree; they are designed to ride over boulders and other debris (this not include intrusive methods such as scallop dredging). UXO already buried is unlikely to be moved by this process and it is very unlikely that even modern EO deposited as the result of relatively recent ad hoc naval and air exercises in the area will be caused to move.

The potential for UXO to be initiated if encountered during project operations will depend on its condition and the energy with which it is struck or moved, or if it is subjected to crushing, friction, static electricity or excessive heat. The movement of vessels and implementation of non-intrusive surveys will not result in the initiation of explosive ordnance through influence alone. In other words where an item is severely corroded, covered in concretion or its batteries are flat, the firing mechanism will not operate as its designers intended.

The UXO could be caused to detonate several ways: if the detonator is struck accidentally with sufficient force or is subjected to heat, static charge, friction or crushing; if a fuse containing a temporarily jammed cocked striker is jarred and the striker is released; similarly if a seized clockwork mechanism restarts; or if the sensitive iron picrates associated with a picric acid filled munitions are subjected to friction, heat or are knocked, particularly if they have been allowed to dry out. In addition to the danger of iron picrates, some other HE can exude metallic azides and salts that, once they dry out, are extremely sensitive. These salts are often hidden within fuse pockets and not readily be seen. An item of UXO may be in a sensitive state, with movement across the seabed sufficient to cause detonation. This movement could be caused by an anchor, cable or wire dragging the UXO. In shallow water (less than 10m), the wake or shallow water suction effect between keel and seabed could be sufficient to move UXO without actually touching it.

The main mechanisms, therefore, that have the potential to cause unintended detonation of an item of UXO are:

• Crushing of the casing, imparting energy to the EO’s detonator leading to its detonation (the main filling is unlikely to be initiated independently);
• A blow with sufficient energy by heavy equipment, dropping the item or, perhaps, a rock against a sensitive fuse pocket or exposed detonator; and
• Sympathetic detonation caused by another item of UXO sufficiently close by or by a shock wave with sufficient energy imparted by an activity such as percussive piling.

Small items of UXO, such as AA, naval and artillery projectiles and small air-dropped bombs, are relatively thick-cased and are considerably more likely to be pushed into the soft sediment of the seabed than crushed (this is obviously not true for outcrops of rock where the sediment is very thin and the underlying surface is hard). Other than in unusual circumstances on hard rock, the likelihood of a detonation via this mechanism for these types of EO is low. Larger naval weapons, such as depth charges, sunken buoyant mines and ground mines have thinner cases and are therefore more likely to be susceptible to crushing.

The leg of a jack-up barge or a vessel grounding during cable laying, together with any associated anchor deployment, have the potential to crush the casing of an item of UXO and shock the sensitive detonator within; even if the fusing system of the EO is no longer able to function as intended due to corrosion or lack of battery power. A glancing blow from an anchor or cable link against a fuse pocket or fuse, could be sufficient to initiate a detonation, but this is unlikely. A blow to a chemical (Herz) horn could cause a sunken moored (buoyant) mine to function; but the degradation of wiring and internal components by corrosion makes this highly unlikely.

In all but the most unusual circumstances, for a high order detonation initiated by the detonator to occur, the EO needs to have been armed; i.e. the detonator is in intimate contact with the primer and main charge. When UXO is encountered, it should always be assumed that the explosive train is intact: that is, all safety measures have been removed and the detonator is in contact with the main charge, although that may not actually be the case. In all cases, encounter and interaction with the UXO must occur first.

Friction and heat are evidently much less likely to cause a detonation underwater than impact or movement. However, it is possible for a small item to become wedged in the flukes of an anchor, PLGR grapnel, cable plough or other equipment and be raised to the surface. In such an event, if the UXO was then subsequently allowed to dry out, sensitive salts (picrates and metallic azides) that had exuded through fuse pockets or corroded shell casing could be very sensitive to heat and friction.

It can be seen that for a detonation to occur, the UXO must be in a sensitive state and a certain set of conditions satisfied. It is evident from the many items of UXO that are recovered from building sites, farmers’ fields, anchor flukes, fishing nets and dredger suction heads by Royal Navy EOD teams every year that these conditions are hardly ever met and an accidental detonation is very unusual. The main filling of most EO is inherently stable and a detonation is a rare event, even when UXO has been subjected to robust handling, for example when a bomb is caught up in a dredger head or ship’s anchor. Most UXO – particularly EO that has lain on the seabed for several decades – will have been the subject of significant corrosion to its casing and to any mechanical moving parts. It is extremely rare for UXO found on the seabed to function as intended; detonation will almost always be the result of unusual and vigorous kinetic stimuli.

Ordtek’s firm view is that the purpose and set up of surrogate item trials should be considered carefully before they are conducted; especially with regard to how the results will be used on a project.

In 2004, the US Army Corps of Engineers, Engineer Research and Development Centre produced “Guidelines for Planning Unexploded Ordnance (UXO) Detection Surveys”. Within this document they stated that “The condition of the UXO also is a factor that influences the ability of a sensor to detect it. Since most ordnance items are comprised of ferrous metal, they deteriorate and rust over time. The measured signal response of a corroded UXO is smaller than that of one in pristine condition”. In other words, UXO can become corroded and harmonised in the local magnetic environment, which in reality reduces the magnetic anomaly strength.

We acknowledge for safety reasons, live munitions cannot be used for geophysical prove-outs. De-militarised inert munitions are preferred. However, even when inert munitions are available, it is important to be aware that the geophysical response of different models in the same munitions class (e.g. 50kg UXBs) can differ. If no actual munitions are available, then surrogates of approximately equal size, shape, mass, and material composition of metal components can be fabricated. It must then be established how much the surrogate’s geophysical response differs to that of the actual munition item being replicated.

However, the fundamental issue with the surrogate items is that they provide a “false” magnetic representation of historic munitions, which cannot be relied upon for target modelling purposes. For example, rolling sheet metal and welding introduces an increased magnetic susceptibility into the item. The danger is that reliance on the elevated results of the surrogate trial could mean that potential UXO related targets are missed during data interpretation.

Ordtek is not suggesting by any means that surrogate trials are worthless and therefore should not be conducted. If properly specified and controlled they can provide confidence that equipment and sensors are functioning correctly and provide an audit trail. However, it is possible that other inherent QA/QC procedures and test pieces could also fulfil this function.

For geotechnical investigation (GI) activities on most sites, Ordtek believes the pre-mitigation UXO risk is generally already low and very close to ALARP, although it is necessary to provide evidence for this stance for a wider industry that is familiar with and “expects” a geophysical survey ahead of all seabed interaction.

It is essential that a full UXO risk assessment (RA) is conducted before confirming the requirement for a costly UXO-specified geophysical survey ahead of a geotechnical investigation campaign. The RA, among other considerations, will determine the likely presence and density of UXO types and an assessment of the environment. The depth of water, for example, is a mitigating factor: deep water will significantly reduce the health and safety risk to personnel and vessels on the surface for most UXO.

The GI activity footprint is usually very small as a proportion of the volume of the site as a whole. Therefore, depending on the assessed likely density of UXO, the probability of encounter with an item of UXO will usually also be very low. It follows that the likelihood of an inadvertent UXO detonation will be even less (see “Myth 2” above). Therefore, in all but the most unusual circumstances, the pre-mitigated UXO health and safety risk to a GI campaign is generally very close to ALARP and the extant residual risk can be mitigated sufficiently – to below the ALARP threshold – by reactive and procedural mitigation measures alone. The cost of a full geophysical survey is usually unwarranted and “unreasonable” within the tenets of the ALARP principle. The risk to equipment on the seabed is, of course, slightly higher and the developer (with Ordtek’s advice) must decide whether this very low Project risk (cost, delay, reputation) is tolerable or requires further mitigation.