Observations and inferences – Reflections on the AVS Database of Demining Incident Victims
Andy Smith, December 1999

The following comments are my own observations and opinions derived from the data compiled in the AVS Database of Demining Incident Victims. They do not represent the formal view of any group or individual other than myself. They are in three parts with these headings:
1 General observations
2 Inferences
3 The revealed protection needs

In the AVS database of demining incidents, injuries are classed as either Severe or Minor. Injuries likely to be life threatening, to require surgery or to result in permanent disability are rated as Severe. All others are rated as Minor. This distinction is for convenience and is not intended to reflect on the discomfort and/or hardship associated with the injury.

The opinions expressed here are made from an informed standpoint but may not be wholly objective. Please check the incident data before making any important decision based on and of these observations.

In some cases, injuries were not recorded in detail but I have resisted the temptation to infer detail wherever possible, and have noted any inferences that occur. Note that, in very severe injuries, only the most severe injury is generally recorded.

1. General observations

The first release of this database contained records of 319 victims and 249 incidents, the majority of which are reasonably well detailed. It was hoped that future releases would include more records, and that further detail will be added to some of the existing records. If your copy of the database has more than 319 records, the figures quoted below will probably vary.

1.1 Injuries

A fragmentation jacket or apron of some kind was issued to under a third of the victims. It was only recorded as worn in half of those cases (but may have been in others). In 17 of these, the victim died. In more than a quarter of those incidents where a deminer was wearing body protection, he still died. The mine-type that most commonly kills its deminer victim is the bounding-fragmentation type, especially the PROM-1, the OZM 3 or 4, and the Valmara 69. These have claimed 17 lives, and ten of those were wearing some kind of body protection.

There is evidence that fragmentation-jacket and helmet/visor combinations that are not designed to integrate and so do not overlap in normal demining positions allow injury to the throat and face. There is also evidence that short selective frontal protection leaves frontal body areas unnecessarily exposed.

33 victims are known to have been injured by bounding-fragmentation mines. Of these 10 are known or presumed to have escaped serious injury. A further 6 are classed as “not known” for want of further detail, but they lived. The remaining 17 died, and 10 of those were wearing body protection. Of those instances where a victim was wearing body protection and survived a bounding-fragmentation mine, the victim was a “secondary” victim and the deminer who initiated the device was killed.

There is compelling evidence that the body protection currently issued is not capable of protecting deminers against a bounding-fragmentation mine detonating at close-quarters. This observation is not new, having been made repeatedly by informed observers, but for the first time it can be taken as “proven”.
The records contain details of 38 deaths. 17 victims died as a result of a bounding-fragmentation mine detonation. Three died disarming (non-bounding) fragmentation mines. One died handling an AT mine. One died as a result of what was probably an IED detonation. One died after a grenade detonation. One died from an unrecorded device that was probably a fragmentation mine. One died after initiating a non-bounding fragmentation mine. The remaining 13 died as a result of blast mine detonations. Four of these were wearing frag-jackets of some kind, but all four were not wearing head protection (or not wearing it properly) and three of these involved severe head-injury. The fourth was squatting and stepped on a mine so suffered severe lower body injury. The Frag-jacket did not “fail”.

There is compelling evidence that the current standard of body protection is sufficient (or more than sufficient) against the blast-mine threat.

Incidentally, there are a higher proportion of fragmentation mine deaths in Bosnia-Herzegovina than any other Humanitarian Demining theatre on record.

There are records of 225 AP blast mine victims in the database, of which 13 (or 6%) died. Nine of the dead victims do not appear to have been wearing facial protection (or wearing it properly). Of the others, at least 48 were not wearing any eye protection. In a further 82 the wearing of facial protection was not recorded. In 46 cases severe eye injury was recorded. In a further 31 minor eye injury occurred. So eye injury occurred for 77 of the 225 blast mine victims in the database (or more than one in three).

For the whole database (covering all device types), the following injuries are recorded (at least this number occurred since many are not detailed):

Severe eye 59, minor 36
Severe face 18, minor 97
Severe head 17, minor 15
Head = 94 severe injuries

Severe hand 32, minor 79
Amp hand 5
Amp finger 22
Severe arm 24, minor 63
Amp arm 9
Upper limb = 92 severe injuries

Severe leg 38, minor 88
Severe foot 8 (+ 6 later amps), minor 10
Amp leg 55
Amp foot 7 (+2 later amp legs)
Amp toes 1
Lower limb = 109 severe injuries

Severe body 12, minor 35
Severe chest 17, minor 37
Severe genital 11, minor 5
Trunk = 40 severe injuries

The differences between the threat to the head and upper limbs is that between 94 and 92, which is not statistically significant in a sample of this size. The jump to 109 for the lower limb injury is significant, and illustrates that the Missed-mine risk is real and generally results in a severe injury. The large drop to 40 for trunk/body injury is also significant, illustrating clearly that the main torso is not at risk of serious injury to the same degree as the limbs and the head.

The trunk/body is the area most frequently protected after the head and the data indicates that this may not be an appropriate priority.

Hearing damage is recorded in some incidents, especially those occurring in Afghanistan. It does not occur with anything like the same frequency in other theatres. It seems that difficulties in assessing minor hearing damage may allow false claims to be made.

1.2 Incident types

Each incident recorded in the database has a classification attached to it. In a few cases these are slightly stretched to fit, but overall the classification system worked well as a general guide to the activity under way at the time of the incident. The classifications are explained under Read me and Definitions at the database opening screen.

114 victims were injured in Excavation incidents
83 victims were injured in Missed-mine incidents
30 victims were injured in Handling incidents
23 victims were injured in Victim inattention incidents
18 victims were injured in Detection/tripwire incidents
16 victims were injured in Survey incidents
12 victims were injured in Vegetation removal incidents
10 victims were injured in Other incidents
7 victims were injured in Demolition incidents
6 victims were injured in Detection incidents

There is considerable variation of the incident types in different demining theatres. Try searching on a combination of incident Class and Country for an interesting breakdown.

The most common activity at the time of an incident is “Excavation”. The next is simply walking and treading on a “Missed-mine”. The next is a mistake when “Handling” and the next an example of human-error, such as slipping or falling over.

Revisions to the techniques used in demining may be expected to reduce Excavation and Handlin g incidents.

Revisions to field discipline may be expected to reduce Missed-min e and Victim inattention incidents.

1.3 Reducing the number and/or severity of injuries

The reduction in number of incidents is only one may to reduce severe injury. Protecting deminers against the effects of detonations is another. Body armour may have limited utility, but there is ample evidence to suggest that the use of long handtools made to appropriate designs reduces upper limb injury, and some evidence to suggest that blast-boots may have some value against lower-limb injury. (I personally doubt the latter.)

While Excavation incidents occur when a deminer is carrying out an activity that must be done, Missed-mine incidents should never happen. Improved tooling and techniques may reduce the incidence of Excavation incidents and reduce the severity of injury when they occur. Improved management and the deployment of appropriately mixed demining methods could be expected to reduce the incidence of Missed-mine incidents.

Long, appropriately designed handtools have apparently reduced upper limb injury when they are used. Long inappropriately designed tools – usually designed for another purpose – have occasionally broken up in a blast and made injuries worse. Search on the relevant entries under the Notes section of the database to find examples of both.

It may be appropriate not to put all effort into trying to stop an incident occurring, but to put some into trying to stop it resulting in severe injury.

 76 of the victims in this database are known to have returned to work. A further 62 suffered injuries so light that they are presumed to have resumed economic activities. In 126 cases, the outcome for the victim is unclear. In only 17 are they known to be unable to work, along with the 38 who died.

More than 43% of the deminers injured in the incidents recorded in this database are likely to be back at work.

1.4 No evidence

.4.1 Blast-damage

 There is no evidence amongst this data for the kind of internal lung injury claimed by at least one internationally active body-armour manufacturer. That manufacturer produces body armour with unique overpressure protection (to prevent internal chest damage caused by the pressure of wave of the gas expansion). The evidence in this database proves beyond reasonable doubt that such protection is entirely unnecessary.

The data in this database shows that, with an AP mine, the major risk of damage from blast is when a part of the victim is within 30cm of the device. The risk is significantly reduced with greater distance.

1.4.2 Blast boots

There is as yet no evidence to suggest that blast-boots would reduce injury in most humanitarian demining theatres. The blast-mine risk in Bosnia-Herzegovina includes the abnormally small PMA-3 (35g Tetryl), but also the larger PMA-2 (100g TNT). Incidents have occurred with both mines with about the same frequency. Current evidence suggests that wearing blast-boots when stepping on a blast mine with significantly more than 50g HE inside it may actually worsen the level of severe injury incurred. Also that the only boots with some effectiveness against the smallest mines include a stand-off of at least 10 linear cm in their design, and these boots would be impractical to walk on in many environments.

It can also be argued that deminer should never miss a mine – and that providing protection that assumes they will miss mines may encourage carelessness.

Many armour manufacturers include blast-boots (or over-boots) in their catalogues that have no proven benefit and may even make injuries worse when a wearer steps on a mine.
I have not seen the results of the recent US Army CECOM NVESD blast-boot trials, and may revise my opinion when/if I am eventually allowed to do so.

1.4.3 Back protection

There is no evidence that wearing a helmet or a back-panel to body armour has ever significantly reduced the severity of an injury.

…………………………………………………………………………… .

2. Inferences

2.1 Deminer management

There is frequent evidence of deminers who are supposed to be “controlling” a partner in a two-man team doing nothing to prevent the partner from breaching SOPs. Self-management in this way does not work. Most of the more forward-looking demining groups are now moving to one-man teams and are appointing more Field supervisors to maintain discipline.

2.2 Field control inadequacies

 In many cases, Field supervisors accept little or no responsibility for the disciplined behaviour of deminers. In some they order the deminer to work in an unauthorised or unsafe way, in others they simply allow it. In many cases they are unaware of the SOPs they are meant to enforce.

187 of the injuries that occur in the database have been assigned Field control inadequacy as their Primary cause . The reason for each is given in the Researcher comment part of each record.

2.3 Management control inadequacies

Many of the Technical Advisors in humanitarian demining are inexperienced in anything relevant. Worse, some of the Program Managers have neither appropriate management skills for man-management in developing countries (not an MBA please!) nor knowledge of the demining profession. The consequence of this is sometimes the deployment of inappropriately prepared, equipped and disciplined deminers and Field supervisors.

81 of the injuries that occur in the database have been assigned Management control inadequacy as their Primary cause . The reason for each is given in the Researcher comment part of each record.

An Afghan group carried out An in-depth study into the demining accidents (METP, 1997 – now META) and identified poor management and supervisory skills as “major factors leading to demining accidents” (p6). (In this case “management” appears intended to mean what I call “field management”). Under “Preventative measures” they listed first improved and more democratic management, second reduced competition between the commercial groups established with UN support to operate there, and third reducing the pressure for deminers to work quickly. The report went on to identify most of the failings thrown up by this database and included a recommendation to issue the 5mm full-face visors favoured by other groups. While the UN have listened to the critique, some of the findings remain to be acted on two years later. Contact Younus Mohammad at META for more information about a little advertised paper that is well worth reading.

Most demining incidents are the result of inadequate management at one level of another. Relatively few appear to have been “unavoidable” (15%). From this it follows that the most effective way to reduce the number of incidents would be to improve management at all levels.

2.4 Reporting incidents

Incident investigations made by Technical Advisors to post-war programmes vary in quality and content dramatically and are rarely comprehensive or based on informed experience. The quality of reports in any one theatre often varies with the frequency that Technical Advisors are replaced.

A “template” for reporting incidents may be useful, along with relevant preparation for Technical Advisors. The format used by MACA Afghanistan is the best I have encountered.

2.5 99.6% Clearance

99.6% allows for the odd device to be missed. However, it is worth considering whether any of the methods currently deployed seem likely to be more efficient than others. Mechanical clearance has yet to achieve respectability, and is not used by any responsible HD group without following up with another trusted method. Other methods, such as burning-off undergrowth are used in preparation but still require a trusted method as a follow-up. The two methods trusted most are the use of explosive-sniffing dogs and manual demining (I recognise that a wide range of techniques are used in both methods).

Instances occur where a deminer steps on a mine in an area cleared by dogs: therefore dogs miss mines.

Deminers from all kind of organisation (commercial, UN and NGO) have been injured by stepping on mines they or their colleagues have missed as they work: therefore manual deminers miss mines.

 The numbers of incidents involving mines missed by dogs in the database is low, but that may merely reflect the fact that globally dogs are used far less than manual methods. However, the database does not cover civilian injury in areas declared clear. There is some evidence to suggest that a study of this would throw light on the effectiveness of dogs.

2.6 Mechanical means

No incidents recorded in the database occurred when manual deminers are following up a vegetation cutting mechanical preparation of an area. I infer from this that mechanised vegetation cutting does not seem to add to deminer risk – and may reduce it.

Incidents do occur in which deminers are severely injured when following up “robust” mechanical preparation of an area. “Robust” in this instance means deliberate detonation of as many devices as possible and not necessarily efficient vegetation clearance. I infer from this that “robust” preparation may add to risk and should be examined critically.

2.7 Eye protection

Given that some lower-leg amputations do not prevent the victim from working again, the most common severely and permanently disabling injury in demining is blindness. This occurs with small blast-mines when inadequate protection (or none) is worn.

I believe that the standard of eye protection offered to deminers is often so low as to render the employer liable for prosecution for criminal negligence. I further believe that the UN’s peculiar legal status (seemingly above the law) makes some UN organised, sponsored and controlled demining the least responsible in the world.
The UN supported initiatives in Cambodia and Mozambique still provide safety spectacles intended for workshop use – despite having been pilloried for this over the years within the industry (at least by me!). (Both are now considering changing to 5mm polycarbonate blast visors.) The UN in Afghanistan has been using 3mm visors and are only now preparing to change to the 5mm blast visor. One specialist NGO that makes a loud noise about using helmets and visors issues its field supervisors with inadequate safety spectacles.

Most deminers do not have adequate protective equipment for their eyes, yet most deminers wearing visors in incidents are not wearing them “down” or “closed” : I infer from this that the eye protection currently on offer is not meeting deminer’s needs.

 Field research shows that full-face visors are often unpopular because they are hot and heavy (compared with wearing nothing). More surprisingly, field research shows that deminers often raise their visors because they cannot see through them properly. Group management at all levels is responsible for ensuring that protective equipment is in serviceable condition – and a visor that cannot be seen through is not worth issuing.

Polycarbonate scratches very easily and is rapidly clouded by fine scratches if it is not protected when not in use. Many groups do not have an intelligent regime for visor-care and many issue visors as a common resource – rather than giving an individual responsibility for the care of his own kit. While these management failings may have contributed to the problem, even groups that do have visor-care regimes in place have difficulty making their deminers keep the visor down at all times.

It is possible that 5mm polycarbonate eye protection that does not cover the entire face may be more likely to be worn properly than the existing visors.

I believe that spectacles or short visors of this material that cannot be raised should be devised and made available as a “user-determined” alternative to full visors. If they were always worn, the incidence of permanent blindness would be reduced.

2.8 Compensation

Deminer compensation varies widely around the world.

In Cambodia the sums paid out for the same injury have dropped dramatically over the last seven years. Also in Cambodia, the main UN backed initiative has penalised deminers for being in breach of SOPs at the time of an incident. At least one seriously disabled deminer received no compensation at all despite having paid into a fund from his own wages. Others have had their compensation reduced as punishment for breaking rules (which their field supervisors did not enforce).

In Afghanistan, the real value of compensation was effectively reduced by at least 50% between 1990 and 1998 when an increase was made that partly redressed the loss. It is paid in Pakistani Rupees, which have devalued against the US$ dramatically. The UN MAC employs a doctor to pursue claims and ensure fair treatment – which is a praiseworthy policy. The doctor acknowledges that past claims for severe hearing loss were frequently unfounded and has made every effort to improve testing and ensure that the fraud stops. He agreed with me that the high incidence of reported hearing loss in Afghanistan should be discounted when assessing the real risk of permanent hearing loss.

In Bosnia-Herzegovina, some major donors gave demining contracts that stipulated insurance conditions. These were poorly thought out and resulted in severely disabled deminers receiving no compensation because they did not actually lose a limb. Fortunately, one commercial demining group taking the contracts felt morally obliged to fund some compensation at its own expense. It seems that “employer liability” is not widely recognised in this industry. While some organisations, NGO and commercial, have a very good record for re-employing disabled deminers, this is done out of patronage rather than legal commitment.

With the exception of settlements to Western European Technical Advisors injured in demining, no settlements come close to meeting the victim’s long-term disabled needs . None known of include a realistic pension potential (although some informal arrangements in Afghanistan approach this).

It should be remembered, for example, that the cost of good health care in Angola is far higher than in most of Europe and the USA, yet settlements are a tiny fraction of what a Westerner would demand.

2.9 Institutionalised lies

Loyalty to a demining programme or fear of losing an income makes some Technical Advisors perpetuate well-known lies about their programmes. The most glaring examples I know of are in UN supported initiatives. Formally, many programmes claim that their deminers lie prone to excavate detector readings. Informally, when faced with the evidence of their own injury records, senior management acknowledge that this is (at least sometimes) untrue. A few of their Technical Advisors are honest enough to refuse to pretend at all about this issue, but others argue strongly that it is the SOP, so it is true.

Search under the Note section of the database for those records where the “victim lying prone(?)” is signalled. In these instances, it is merely possible that the victim was lying down, not certain. Deminers do not routinely lie prone to excavate.

Pretending about something as fundamental as the stance in which the deminer works is counterproductive – leaving armour manufacturers trying to design for the wrong position and deminers confused about the right way to work. If they can ignore one rule, why not the rest?

A senior manager at one well known specialist NGO claimed they did not keep records of incidents at their UK base. The researcher believes that this is a lie because it would be very irresponsible of an otherwise respected group if it were true.

Another specialist NGO issues armour with back-panels and helmets with visors, claiming that their deminers are the best protected in the world. In fact, their field supervisors in Cambodia are issued with industrial safety spectacles and the armour back-panels with high collars make it impossible for the deminers to work as directed.

The Humanitarian Demining industry needs an independent body that observes objectively, (with some understanding of demining and working in developing countries), and reports honestly on what is actually happening.
…………………………………………………………………………… .
3 The revealed protection needs

 Among many equipment and management needs, the data shows that there is some advantage in deminers wearing protective items when they are worn properly. There is also some advantage in using tools long enough to put the victim at a distance from any AP blast that occurs.

3.1 The threat

I do not define the threat as the mine most commonly found in a particular theatre. I define the threat as the mine(s) most commonly occurring in recorded incidents in that theatre. Some of these are surprising – being relatively easy to detect. Others are the mines you might expect.

3.1.1 Blast

Briefly the data reveals the blast-mine threat to deminers in the various theatres as:
Afghanistan – PMN (240g TNT) mine features in 62 (of 77) injuries
Angola – PPM-2 (110g TNT) mine features in 12 (of 29) injuries.
Bosnia-Herzegovina – PMA-3 (35g Tetryl) mine features in 7 (of 15) injuries: the PMA-2 (100g TNT) features in 5 injuries.
Cambodia – Type 72 (a or b) (51g TNT) mine features in 13 (of 45) injuries: the M14 and MD82B (27/28g) features in 8.
Iraq – the PMN (240g TNT) mine features in 5 (of 7) injuries
Laos – none recorded
Mozambique – PMN (240g TNT) mine features in 14 (of 24) injuries
Zimbabwe – R2M2 (58g RDX/WAX) mine features in 6 (of 8) injuries
In all countries except Bosnia-Herzegovina and Laos, the PMN and/or PMN-2 feature among the mines involved in incidents.

3.1.2 Fragmentation

Briefly the data reveals the fragmentation-mine threat to deminers in the various theatres as:
Afghanistan – POMZ (75g TNT) mine features in 10 (of 13) injuries.
Angola – POMZ (75g TNT) mine features in 1 (of 1) injuries.
Bosnia-Herzegovina – PROM-1 (425g TNT) mine features in 17 (of 17) injuries.
Cambodia – POMZ (75g TNT) mine features in 1 (of 1) injuries.
Iraq – the Valmara-69 (450g Comp B) features in 3 (of 3) injuries (the PROM-1 also features in 2 of these)
Laos – a mortar features in the only recorded injury
Mozambique – OZM-4 (170g TNT) mine features in 7 or 8 (of 9) injuries
Zimbabwe – none recorded
The PROM-1, OZM-4 and POMZ represent the greatest threat (in that order) but the PROM-1 does not feature in the data for Cambodia, Afghanistan, Laos, Zimbabwe, Angola or Mozambique.

One striking conclusion is that the threats in the former Yugoslavia are different from those elsewhere, which implies that the level of protection may need to be different.

3.2 Protecting against fragmentation

Protection made to reach a STANAG V50 of 450m/s has proved less than adequate against fragmentation mines. This was, sadly, predictable.

The expression “V50” means that at the measured speed, 50% of the fragments may penetrate. The STANAG fragments used are hardened steel, shaped like a blunt chisel and they must hit the target squarely, so are usually fired down a rifled barrel. They must strike the target in a predetermined pattern and the test is conducted under rigorous climactic conditions.

It hardly needs saying that no fragmentation mine ejects straight-flying chisel-shaped fragments one at a time at measured spacing in controlled climactic conditions.

In a real situation using a large bounding fragmentation mine (OZM-72), a spray of ragged, very hot fragments were ejected at speeds varying (based on empirical tests) from below 200m/s to over 1200m/s. At close range, they not only strike the target at effectively the same time and often very close together, their impact is also closely followed by a blast-pressure wave. Even in my tests with a POMZ a wide range of fragment speeds (some over 600m/s) was observed.

The current UN body-armour standard of a STANAG V50 of 450m/s, published after the Copenhagen conference when “experts” decided on protection needs, does not even guarantee protection against the smallest common fragmentation mine (the POMZ). It has failed repeatedly to protect against the PROM-1 in Bosnia-Herzegovina.

Despite this one armour manufacturer (KE BURGMANN) sells 450m/s armour to deminers in Bosnia-Herzegovina with the label emblazoned “Fragment proof demining vest”. I photographed a blood-stained example in June 1999.

The STANAG V50 is not a relevant test for demining fragmentation armour. Another, using combined multiple fragmentation and blast threat, should be devised.
Most fragmentation-jackets in use were not designed for demining but for police or combat situations – along with the standard they are measured against. Humanitarian deminers do not need to hold a gun in both hands, they do not need back protection, or to run crouched into a low-profile while wearing it. There are some purpose designed armours around, but many are poorly designed and made.

The adoption of another existing standard might seem attractive. I advise anyone thinking this to read Jason & Fackler’s Body Armour Standards – a review and analysis (published by the US Center for Ballistic Analysis in 1990). It shows clearly how none of the existing standards are thorough, and none apply to the demining threat.

As the data shows, the armour currently issued is not always worn. Deminers tell me that this is because it is heavy and uncomfortable and they feel it may make it more likely that they suffer an incident. The deminer’s view is all important. He is not supervised at all times (as the data shows) and his views must be respected. For this reason, I do not believe that the simple solution of making deminers wear bomb-suits is likely to have the desired effect of significantly reducing injury and death. I do believe that such a policy would slow down demining and lead to more civilian casualties while they wait for the deminers to arrive.

There are other ways of dealing with a bounding fragmentation mine risk – methods that keep the deminer a good distance away or behind solid armour. All bounding mines that I know of are laid with the trigger part of the device exposed. By definition (containing a high volume of metal fragments) all are relatively easy to detect. Many of the injuries occur after detection while trying to render the device safe. Strategies to deal with these mines by deliberate detonation during mechanical preparation of the ground are already being developed. Strategies to deal with them where detonation and widespread fragment dissemination are undesirable need to be developed. (Detonation inside a wide diameter re-useable armoured tube – possibly with added water – should be explored.)

Assuming that there is a need to act now rather than following a long period of R&D, the only way to realistically protect against the fragmentation threat is to develop strategies to avoid it.

3.3 Protecting against blast

 The main mine threat in humanitarian demining is the AP blast mine. I will not discuss the risk when standing on a mine, except peripherally in terms of desirable “stand-off”.

When searching for and excavating AP blast mines the threat is not simply from the “blast-wave” velocity, but also from environmental fragmentation from the earth and debris surrounding the mine. In a few cases, fragments of mine cases have also caused injury. A grain of sand that would “sting” the face and leave no permanent scar can cause permanent blindness if it hits the deminer’s eye. The “blast-wave” itself causes severe damage to parts in very close proximity.

Objective and thorough research into the effects of detonating relatively small quantities of high explosive close to a person are hard to come by (impossible). To some extent this is because the variables in each detonation and with each potential victim are large, so giving a definitive answer is difficult. Researchers tend to err on the side of safety and predict a range of outcomes with a large safety measure. Theoretical outcomes must assume “optimum dispersion of oxidiser and fuel elements… a situation that rarely exists in practice” (R1). Other researchers, with a commercial interest, are not above blatently twisting facts to suit their desire for a market niche and profit.

The events in an HE blast have been studied at length over the years, mostly with regard to much larger blasts than those associated with AP blast mines. The oscillating blast-wave front and peak-overpressure risks are well documented (if not often intelligibly) alongside assessments of reflection from buildings and predictions of structural damage. While this research has often included an element of empirical verification, it is on a scale that dwarfs AP mine blasts. The results are often extrapolated downwards to assess AP mine risk without verification on that scale, and without using the correct explosive type of a representative age and condition to confirm the theory.

The only way to verify the theoretical risk published by researchers is with empirical tests. In addition to examining the injuries and damage to equipment in this database, I have carried out about 50 controlled detonations of landmines (of all kinds) and presented various materials and equipment to the blast in ways that mimic the position of a deminer detonating a device. As a result I find my opinions do not match those of any theoretical scientist. They do come close to the published results of H.J.Yallop in the forensic science text-book Explosion Investigation as long as allowance is made for his addition of a “safety factor”.

Yallop says:
The detonation of 100g TNT would create a pressure of around 207 kNm -2 .
The detonation of 200g TNT would create a pressure of around 391kNm -2 .
He reports a high chance of eardrum damage at pressures between 2-300 and a slight chance of lung damage. A real risk of lung damage would not occur until closer to 500 kNm -2 .

For many readers “kNm -2” will not mean a great deal. Think of it as the force behind the advancing blast-front. The speed of that front (for TNT) is recorded as 6,825m/s -1 in a technical briefing on Explosives Detection that came my way from AI security of London. The speed starts off the same with the same explosive no matter how big the charge, but slows down more quickly when there is less force behind it (when the charge is smaller).

It is important to remember that the blast-wave of a detonation is expanding. The volume behind the wave doubles in the same period of time (part of a millionth of a second), but as the “front” of the expanding “ball” gets bigger, it actually advances very much more slowly. The pressure is spread behind an ever increasing surface and this means that the pressure at the “front” of the wave is very reduced even by a small distance from the blast source. (It seems that the expanding wave bounces away from its own front, returns to the source and spreads out again. It “oscillates” or “echoes” with secondary waves like the ripples on a pond, but this happens so quickly that it is not relevant to us.) Many of us intuitively know that we are safer even a small distance further away from a blast. This merely adds theoretical weight to something that is empirically obvious – and supported by the evidence in the database.

What is relevant to us is that a deminer using a 15cm bayonet to expose a PMN and who lets it off has a high risk of the blast causing severe hand damage or loss. The same deminer using a 30cm (or longer) bayonet stands a much reduced chance of suffering severe damage. (The bayonet is not an ideal tool because the handle fragments, but the point about increased distance is supported by evidence in the database.)

At the time of detonation, the fragments associated with the blast are either travelling at the same speed as the blast-wave/front or slower than it. A very short time later those fragments are moving ahead of the blast front.

When a deminer wearing a full-face 5mm polycarbonate visor prods onto a mine, his visor is struck by the environmental fragmentation before the blast-wave/front reaches him. The visor face is marked by the fragments and then the visor is often torn off as the front passes (this even occurs with helmet/visor combinations when the helmet is not strapped in place). Evidence of this is found among the incident data. I have carried out more than a dozen empirical tests and found that at a 60cm distance from a PMN (240g TNT) with a 75g detonation charge (over 300g in all) the visor is always first struck by the fragments, then by the blast wave. This observation is supported by evidence in the database where in some incidents deminers lost their visors without suffering facial damage.

As the pressure front passes there is negative (or near-negative) pressure behind it. In practical terms, the expanding wave creates an in-rush of dust behind it and the deminer is often temporarily blinded by the dust – but as long as the visor was down his eyes are not damaged.

There is ample evidence to suggest that 5mm polycarbonate protection, worn properly, protects the eyes against all the blast mine risks so far encountered in Humanitarian demining. It has a STANAG V50 of 250-280m/s (depending on who conducts the tests – my own result in 279).

This implies that the published (and entirely unwearable) standard for eye protection of 450m/s is rather too high. To achieve 450m/s requires a 12 or 13mm polycarbonate visor which has been universally found to weigh too much for sustained wear.

It further implies that the 450m/s body armour standard may be much higher than needed to protect against blast threats. When increased protection means high cost, greater weight, higher discomfort and a real chance that the protection will be discarded as soon as possible, there is a need to make the standard realistic.

Bearing in mind that deminers do not really lie down to prod (wouldn’t honesty be convenient! – DRES in Canada wasted a lot of timing testing a suit in the prone position in 1996 – and in September 1999 the MAC for Afghanistan released a Pick-SOP that assumed the tool would be used lying down!) the deminer’s body is closer to the centre of the blast than his face, so the blast wave will be travelling faster when it strikes his thighs or body. How much faster is hard to guess and probably impossible to calculate accurately.

I made aramid armour with a V50 of 380m/s. In all honesty, I did not understand how the fragment stopping velocity related to the blast-wave pressure and simply used it as a rule of thumb that allowed a comparison between existing equipment.

The armour with a V50 of 380m/s passed multiple empirical tests against PMNs and went into local production in Africa. It has so far sustained more than a dozen blasts in incidents, some at less than 20cm (four were held in the wearer’s hands!) without significant damage or penetration. Even when the mine was within 20cm of the deminer’s chest, there was no evidence of internal chest cavity damage. Either the armour significantly stopped it or the blast-wave/front was not travelling at a damaging speed. (In those cases the mine had only a 58g RDX main charge.)

It is possible that if the deminer were against an immovable object (like a large tree), the wave would pass through him causing damage. In the recorded cases where a deminer detonates a mine while excavating, the blast wave/front appears to knock him backwards – applying pressure lower on his body momentarily before his head so knocking him straight back so that he then falls over rather than pivoting him as though his feet were fixed to the ground. In several cases the deminer is lifted up and carried backwards before being dropped to the ground. There are no records of the victim being injured by striking the ground after the impact. Strangely, this is even true of cases where deminers step on mines and are often reported to be thrown into the air. (In one case a deminer walked away confused after detonating a mine with his prodder. He then fell flat on his face and cut his lip, but even this is rare.)

This implies that armour with a significantly reduced protection level – so cheaper, lighter and potentially more comfortable – would be adequate for the blast-mine risk.

 It should be stressed that the addition of a large “safety factor” that ends up making the protection unworkable is counter-productive.

SIZE="+1">In conclusion
A workshop on “Anti-Personnel Mine Blast Injury” was held at National Defence Headquarters in Canada in August 1998. The report of that workshop illustrates intelligent and observant people struggling to understand a threat without many tools to do so. One Canadian with Cambodian experience spoke from an informed base, most did not. One participant thought that “primary fragments [from mines] are relatively easy to measure and characterise”. I wish this were so. A medical doctor talked at length about post-operative complications that smacked of the theoretical battlefield rather than anything seen in this data. Everyone seemed to think modern-medicine was available where it is not and that long-term aftercare could be relied on. Sadly, even in Europe this is not the case. But the report of the workshop is by no means all bad. One speaker observed that deminers would be “willing to give up protection to gain on comfort and appearance” – an unconventional view, but probably true. A speaker for the Canadian armour company that pretends internal chest injury is a major risk was honest about the limitations of blast-boots, which was refreshing. And the need for a new and appropriate testing regime for blast and environmental fragmentation was highlighted – at last!

The workshop concluded with a reiteration of the myth that the torso is the main area to protect, and a call for the Red Cross to provide its civilian mine-injury data. The activity of civilians when injured is not that of a deminer. In most cases, the device that injured them is not identified. It is hard to see how the Red Cross data would be useful. The data in the AVS database is – and could be more useful if those in the field supply additional data for updates.


To discuss these and any related issues, contact me on avs@landmines.demon.co.uk

…………………………………………………………………………………

SIZE="+1">References
R1 Physics of Explosion Hazards – Paper - Bibhu Mohanty: para 2.6

Explosion Investigation , H.J.Yallop, Forensic Science Society and Scottish Academic Press, 1980, UK. ISBN 0 9502425 5 1 – special thanks to Bob Keeley for directing me to it.

Jason & Fackler’s Body Armour Standards – a review and analysis (published by the US Center for Ballistic Analysis in 1990)

An in-depth study into the demining accidents (METP, 1997) MAPA, UNOCHA, Islamabad.

Anti-Personnel Mine Blast Injury (minutes of) National Defence HQ (Defence R&D Branch), Ottawa, Canada, August 1998.

Mine Blast Trauma – experience from the war in Afghanistan – Nechaev et al, Council Communication, Stockholm, Sweden 1995

Forensic Pathology of Victims of an explosion – paper – James A.J. Ferris

Independent report into the demining accidents – E.Banks, Bosnia Herzegovina 1997

Donor influence on safety and productivity in Humanitarian demining – paper – H.Thompson, JICA-UK, August 1998.

Blast injury – paper – World EOD Gazette, September 1998

Blast effects on the human body – A. Purvis – paper World EOD Gazette, July 1998

Testing the . . . Mine Clearance Suit against Anti-personnel Mines – Bergeron & Walker, DRE, Suffield, Canada, July 1996.

…………………………………………………………………………………

Author’s papers (on humanitarian demining issues)

Observations & Inferences – Reflections on the AVS Database of Demining Incident Victims – included with the software, Andy Smith, AVS, June 1999. Published by Mines Action Canada at http://www.minesactioncanada.com/competition/smithmineinjuryreport.html

Database of Demining Incident Victims – Vols 1-3 – US Army CECOM NVESD, Andy Smith, June 1999.

Something is rotten in the state of demining – paper presented at (and in proceedings of) SWEDEC & Nordic Demining Research Forum conference, Sweden, 8-10 June 1999.

Training and certification – paper presented at (and in proceedings of) SWEDEC & Nordic Demining Research Forum conference, Sweden, 8-10 June 1999.

Demining incidents in Bosnia Herzegovina , US Army CECOM NVESD, Andy Smith, June 1999.

Demining incidents in Cambodia, US Army CECOM NVESD, Andy Smith, March 1999.

Demining incidents in Southern Africa , US Army CECOM NVESD, Andy Smith, January 1999.

Demining incidents in Afghanistan, US Army CECOM NVESD, Andy Smith, September 1998.
SIZE="-2">
Standards and Safety in Humanitarian Demining – paper presented at (in proceedings of) conference entitled “Standards & Measures of Success” James Madison University, 4-7 October 1998.
Myths, Mines and Ground Clearance – Journal of Humanitarian Mine clearance, James Madison University, Issue 2.3 October 1998 – http://www.hdic.jmu.edu/hdic/journal/2.3/features/future.htm

The future of Humanitarian Demining – Journal of Humanitarian Mine clearance, James Madison University, Issue 2.3 October 1998 – http://www.hdic.jmu.edu/hdic/journal/2.3/features/future.htm
SIZE="-2">
Current situation and perceived needs for head and face protection in humanitarian demining — in proceedings of “The design and Integration of Helmet Systems” Conference held at the Sheraton Tara Hotel, Framingham, MA, USA on 2-5th Dec 1997: Journal of Humanitarian Mine clearance, James Madison University, Feb 1998 – http://www.hdic.jmu.edu/hdic/journal/2.1/home.htm

There are no deminers at this conference — presented at (in proceedings of) A Global Ban on Landmines: treaty signing and Mine Action Forum, Ottawa, Canada December 1-4 1997.

Developing new equipment for use in sustainable mine clearance programmes — University of Warwick DTU, Dr T.H.Thomas & Andy Smith, 1997.

Introducing regionally manufactured demining visors to Southern Africa — University of Warwick, Andy Smith, 1997.

Sustainable methods for clearing landmines after conflicts — Royal Engineers Journal August 1996 (Vol 110, No.2) p160, Dr T.H.Thomas & Andy Smith.

Surveying demand for demining equipment and proving production capacity in Cambodia — University of Warwick, Andy Smith, October 1996.

Surveying demand and proving production capacity (of protective equipment for mine clearance) in Cambodia — University of Warwick, Andy Smith, December 1996.

Shields and Visors for local manufacture — University of Warwick DTU, Technical Release 40, Andy Smith, 1996.

A demining study in Mozambique: November 1995 — University of Warwick, Working Paper 48, DTU, Andy Smith, 1995.