Issue 6.1, April 2002


National Mine Action: Problems and Predictions


by Vernon Joynt, Technical and Scientific Consultant to the Council for Scientific and Industrial Research (CSIR), South Africa



(Above) Fuel Air gasbag in position for mines among the Pylon legs. (Right) Fuel Air gas detonation initiated with a rifle bullet.

I recently received an e-mail with real good bait on the line so I would commit to sticking my neck in the noose once more. The questions included:

  • Will technology ever improve speed and safety in Humanitarian Demining (HD)?

  • What is the comparative efficiency among commercial, military and NGO clearance?

  • Will dogs be more or less integrated into HD programs?

  • When will land use priorities determine clearance priorities?

  • Is "donor fatigue" a reality? If so, how do we fix it?

These questions are right in the middle of my current pet struggle to give technology its rightful role and place in mine clearance. I would like to discuss both technology in (HD) and donor fatigue, because technology is linked to funding. Present HD methods are too slow and ineffective at solving the total problem, so the donors do not get value for their money.

I want to present this article simply as a South African who has been involved in HD on one hand and associated with research and development (R&D) on the other. By 1996, as one in the South African government-owned demining group, I was already using a presentation slide that read:

A Technologist’s Nightmare: You build the SILVER BULLET and nobody wants to use it.

Little did I realize then that this nightmare would turn out to be the truth for technology in general and not only for one individual in particular. The fact is that new technologies and products are being blown out of the game before they have been given a fair chance to develop into their full field potential. Initially I blamed the people

Steel Wheel Casspirs pulling Disc Rollers detonating AP mines

controlling the demining as being responsible. Now I realize that people developing the technologies are at least equally responsible. Too many are simply doing their R&D for R&D’s sake or for financial gains with too little drive to get the product field-ready for deminers to use. This act in turn is upsetting the deminers in the field, justifying their attitude that the research monies should be used directly for demining.

Will Technology Ever Improve Speed and Safety?

The simple answer is YES. In fact, for us as a demining group, it already has. For example, in 1991–92 we cleared 12,000 mines from around 62 pylons on the Maputo–South African border power line in Mozambique. We completed the job with two Casspir mine resistant vehicles (MRVs) fitted with various demining accessories; two armoured bulldozers; and a detonating fuel-air gas mixture in one-meter diameter plastic-film tubes to work in spots that the machines could not get to, such as amongst the pylon bottoms and stays.

The blast pressure of the fuel air gas mixture is sufficient to set off the functional PMN and PMD6 mines used, but not the PMN2, which is blast protected. Due to the fast movement of the PMN2 pressure plate during the blast, we found the shallow-buried mines would jump out of the ground, making them visible.

AP mine blast pattern after a Bomb Arm with Discs, sweep.

The Casspir bomb arm with discs sweeping AP mines.

Over most of the area being demined, the active mines were detonated by steel wheels and dragged disc-rollers. As a check, the bulldozers would push the soil and dead mines into heaps of earth that were sieved mechanically to limit what we missed. At that point in time (1992), all that the contract required was a safe surface to work on for the pylon repair crews, so only functional mines had to be removed, while UXO and dead mines under the surface were acceptable. Our crew played soccer on the swept areas to prove that no active mines were left behind.

A similar pattern appears after the normal steel wheel and disc rollers had passed. The gaps in the pattern showed where a duff mine was to be found and the deminers then simply destroyed it. For clearing military-laid minefields this pattern is the big advantage of rollers as opposed to flails or tiller machines. The latter leave no such patterns and move around the mines that are not neutralized. Using a backup detection array is also an advantage or even a necessity.

There are many in the demining community who have not really caught on to the South African approach of using MRV’s like Casspir, Buffalo or Wolves linked to Steel Wheels and Rollers to detonate mines while flattening and removing vegetation. Putting in a machine that breaks and spreads the mines, the pieces of which you must later locate, does not support the approach. The choice of starting with a flail or tiller as the first step in bringing technology to mine clearance, as is happening at present, is counterproductive for several reasons. It will make the use of dogs and other detection principles a problem. Broken pieces of TNT mix into the soil, making the use of dogs and other vapour detectors dubious. Techniques like Ground Penetrating Radar (GPR) also need the mines to be intact and upright, and the soil to be uniform without air gaps in it. Also, moving the surface shrapnel into the ground makes using selectivity in metal detection more difficult to do. Now back to the example.

The total area cleared was 96,000 square meters, cleared in five weeks by seven men operating the machines and gas. One man was hurt on the bulldozer when he removed the visor of the bomb suit he was wearing because of the heat.

While de-arming a SPM limpet mine, a full bomb suit, helmet and visor is being worn. These are all capable of stopping the shrapnel from a PROM1 mine.

This bomb suit and helmet was used by the driver of the bulldozer. The helmet and visor are considerably stronger than the types used by deminers. These can stop the shrapnel of a Valmara jumping mine at a distance of one meter. This pre-fragmented shrapnel is equivalent to that of a PROM 1. The only problem with the suit is that the price is $8,000 (U.S.) for the helmet alone! A pellet from a jumping mine had gotten between the driver’s head and the helmet, grazing his head and drawing blood. He was lucky. The UN statistical figures for removing 12,000 mines show that we should have had at least six casualties, two of which could have been fatal.

The cost was $134,000, or $1.40 per square meter and nine cents for each mine destroyed; 450 square meters were cleared per man per day. During later contracts where we worked under UN standards and then had to also lift UXO and dead mines, the figures changed with the addition of manual demining teams who could work on such flat and foot "safe" ground at a rate of 300 square meters per man per day—up to 15 times faster than normal for working in virgin mine fields—so it only raised the cost to $1.50 per square meter. In this first example, technology proved its value by removing vegetation and providing foot-safe ground.

Another example occurred in 1996, during the UNAVEM III contract in Angola, when 4,880 km of road was cleared to a width of seven meters in one year for $6.5 million (U.S.). We cleared 215 mines consisting of improvised AT mines, normal AT mines and AP mines in the shoulders of the road and abutments of the bridges. A further 802 elements of UXO were uncovered. We used a total of 70 men, most of whom were manual deminers hired from a Zimbabwean and a British demining company. They were supported by 24 Casspirs and two Chubby systems run by South Africans and Angolans.

A very important component of the team was the eight mine detecting dogs (MDDs) of the American subcontractor and the 20 South African dogs, half of which were used at Cabango in Angola while the others were sent to Pretoria to run the detection component of the Remote Explosive Sensing Techniques (REST) system we were using for detecting low- and no-metal mines while doing QA behind the manual deminers. The REST system did the area reduction, with the dogs searching out low metal content mines. They found three no-metal mines that the handheld detectors missed completely. The dogs held the key to the speed and cost advantages achieved in the contract, thus proving their worth in HD.

The total area cleared was 34,160,000 square meters (3,416 hectares or 8,440 acres), so the clearance was done at 19 cents per square meter (a third of the cost of normal manual demining), with 1,500 square meters cleared per man per day (at least four times quicker than usual for manual demining).

Today, other demining groups that do not use REST question locals and study other information available to do the area reduction. In Angola, this would be foolhardy because AT mines need a vehicle wheel to trigger their firing mechanisms, and the vehicles using the roads were never using the full seven-meter width. On the contrary, they were often carefully using two-wheel tracks weaving down the center of the wide road. Furthermore, at least half a dozen groups were laying mines at various stages of the war, so information from those sources was not to be trusted even if it was available. The REST system has a scientific basis and has proved to be quite reliable over the course of more than 7,000 km of road contracts. Norwegian People’s Aid (NPA) is presently using REST in Angola for road clearance and have found that they can speed up their operation considerably by doing so.

A third example concerns a contract done in 1998 in southern Mozambique. In this contract, we cleared 1.6 million square meters in less than six months using 28 deminers, removing 5,400 mines in the process. The pricing was distorted because the contract included training dogs, handlers and advanced demining managers for the local Mine Action Center (MAC) as a separate effort in the contract. In a separate calculation, the pricing for the actual demining worked out at 38 cents per square meter. This was at least half the going rate for manual demining in that area. In this contract, we also did some work using a Casspir mounted metal detector array. It proved very successful.

Based on a clearance comparison done on a 20,000 square meter (two hectare) piece of the mine fields between the three-meter-wide detector array on the Casspir and our manual deminers using hand held detectors, the array managed to work 100 times faster, yet it still found all the AP mines that the deminers could find. The mines lifted were a mixture that included PMN, PMN2 and PMD6 mines. These are all low metal content AP mines. An important observation that came out of this test was done while we were weighing and inspecting every piece of metal the hand-held detectors were signaling. We noticed that most of the false signals were pieces of shrapnel and metal junk like wire and bottle tops that were either on top or in the first 10mm of soil. The array had been modified to ignore such small surface signals but to still find a PMN2 on the surface. This was the key advantage of the array.


  • The manual deminers worked behind the array, and they uncovered an additional 1,640 metal signals not marked by the array. The 30 AP mines it did mark were the only mines they could find in the two hectare site, however.

  • The array did mark a further 107 pieces of metal. Therefore, the array marked 15 times less false alarms than the hand detectors without missing the mines. So it was much more selective than the hand detectors.

  • In light of the Database of Demining Incident Victims (DDIV) facts about how manual deminers miss at least three percent of mines, it would have been interesting to have had the array behind the deminers. We may have had a shocking result for the deminers!

  • Arrays have their coils overlapping and cannot leave unchecked spots, something it is suspected manual deminers can easily do.

  • The test was done on a mine field that had its vegetation run flat and made foot safe by Casspirs. The mines found were nonfunctional, probably due to moisture ingression destroying their detonators.

Comparative Efficiency Among Commercial, Military and NGO Clearance Teams

A Mine Detecting Dog looking for mines in a road in Angola.

There really needs to be no choice favoring any of the groups on efficiency, unless financial constraints affect a bias. The way commercial contracts are presently structured, commercial companies are at a disadvantage. The size of the contracts is simply too small to allow the use of expensive technologies to give them an advantage. Militaries can use such technologies and only pay operating costs. Let me illustrate this with figures out of our company’s experience.

In the period 1991 to 1996, we did 11 contracts at an average contract size of just over $1.5 million per contract. Three contracts were over $2.5 million, allowing the use of technology and three-fold increases for price and speed. During the period from then until the end of 2000, we did 28 contracts for which the average contract price was slightly less than $400,000, and there was only one proper contract ($2.7million in 1998) in which we could use the technology effectively. This was the third example already given above.

The succession of small contracts where high capital investment and mobilization cost made the Casspirs stay away virtually turned our group into a dog training company. The REST experience fortunately has given us additional knowledge in the training and use of normal MDDs. This contract size problem can be seen as one of the main reasons that the R&D component of our group has been moved into the South African government’s R&D organization, the CSIR. R&D simply cannot be supported or properly used in these small demining contracts. The situation can be likened to a road building company trying to use its graders and dozers to compete for the repair of suburban side walk contracts.

If any one group must bear the blame for retarding technology in HD it must be those responsible for fragmenting donor monies into these small packages. This really leaves demining technology in the hands of the military. The military has the further advantage of generally

Manual deminers checking the Vehicle array paint marks and re-clearing the whole two hectare mine field.

assembling the best equipment, even if it is quite expensive. NGOs that have visionary management and strong financial support may become the technology leaders in HD. As for the commercial companies, in my opinion they have no hope of competing, as long as the prescriptive nature and size of the contracts remains small as they are at present. The companies are then forced into using manual demining and MDDs, instead of the technologies that have in fact already proven to be more cost effective, faster and safer. So in the end, it will probably have to be the military, which is accused of causing the problem in the first place, that will have to provide the final solution to the problem.

Is "Donor Fatigue" a Reality? If So, How Do We Fix it?

(The following section comes from a paper I wrote entitled Why Do R&D?)

From 1997 onwards, the aid money was always split down into such small pieces that technology-based techniques could not be used. It is my opinion that this was done to favor manual demining in the false belief that it would make area clearance cheaper. What in fact happened was that manual demining made area clearance a lot slower and in fact twice as expensive. This in turn drove the donors away because they were not getting value for their money, and they also saw no end to the problem in any of the contaminated countries. Humanitarian Mine Clearance (HMC) was seen as a bottomless pit into which money needed to be poured for another century or two before the problem was solved! Today everybody wonders about so-called donor fatigue. Even worse, the clearance efficiency of manual demining is now being severely challenged—another shock for the donors.

Instead of HMC, I predict that Commercial Clearance Contracts linked to much larger financial projects would serve to bring high-technology techniques back into Mine Clearance (MC), so we should concentrate on these techniques. Apart from having a larger funding profile, such contracts are run by engineering principles where speed and efficiency are insisted on and properly measured. I am thinking of contracts for building dams and roads, laying pipelines and repairing railways that have mine clearance requirements.

In this context, there has been a recent incident where a survey technique regularly used in a HMC context was used in a survey done by the demining company contracted to do the survey. There was a fatal explosion in an area declared clean. This let the company doing a critical construction contract down. Talk has it that the company involved in the survey will not be used again—the possible financial losses to the larger contract due to doing the survey unscientifically will not allow it. The same problem can also exist in the clearance operations.

Whether the above is true or false, we observed this behavior in 1994 when we were contracted to redo clearance on three so-called cleared roads done by manual demining. We found quite a few additional mines left behind besides the ones that had initially upset the road builders. Men simply put down their tools when their bulldozer finds AT mines in locations that the MAC had declared safe. Obviously, when this type of result-oriented MC becomes the order of the day, then technology and its R&D component becomes more important and more viable. So I put a priority on R&D.


The management of HD must reflect on the fact that the established techniques they are now supporting as the way to solve the big problem are too slow, not cost effective and causing too many casualties. Only when they have successfully addressed these problems will the donor fatigue and other growing problems be solved. Finally, everybody must get much more serious about getting the appropriate technologies into the field.

Contact Information

Vernon Joynt
Technical and Scientific Consultant
CSIR South Africa


Click to learn more about JMU.

  Publisher: MAIC  Contact: 
A James Madison University Website