science and justice - policia

Science and Justice xxx (2015) xxx–xxx

SCIJUS-00580; No of Pages 9

Contents lists available at ScienceDirect

Science and Justice

j ourna l homepage: www.e lsev ie r .com/ locate /sc i jus

Demining Dogs in Colombia – A Review of Operational Challenges,Chemical Perspectives, and Practical Implications

Paola A. Prada a,⁎, Mario Chávez Rodríguez b

a Texas Tech University, Institute for Forensic Science, Lubbock, TX 79414b Colombia National Police, Bogota, Colombia

⁎ Corresponding author.E-mail addresses: [email protected] (P.A. Prada),

[email protected] (M.C. Rodríguez).

http://dx.doi.org/10.1016/j.scijus.2016.03.0021355-0306/© 2015 The Chartered Society of Forensic Scie

Please cite this article as: P.A. Prada, M.C. Roand Practical Implications, Sci. Justice (2015

a b s t r a c t
a r t i c l e i n f o
Article history:Received 15 October 2015Received in revised form 23 March 2016Accepted 24 March 2016Available online xxxx

Within the framework of an internal armed conflict in Colombia, the use of antipersonnelmines by revolutionaryarmed forces represents a strategic factor for these groups. Antipersonnel mines are used by these revolutionaryforces as a mean to hinder the advancement of the national armed forces in the recovery of territory and to pro-tect tactical natural resources and illegal economies within a given area. These antipersonnel mines and impro-vised explosive devices (IEDs) are not of industrial manufacturing, and have a variety of activating mechanismsas well as non-metal materials which make them difficult for successful detection. The Colombian experiencestrongly represents the current need for advanced research and development of effective field operations withinits affected territory. Current efforts are focused on a more operational demining perspective in coca cultivationsites in charge of mobile squadrons of eradication (EMCAR) from the National Police of Colombia working to-wards a future humanitarian demining upon an eventual peace process. The objectives of this review are notonly to highlight already existingmine detectionmethods, but present a special emphasis on the role ofmine de-tection canine teams in the context of this humanitarian issue in Colombia. This review seeks to bring together adescription of chemical interactions of the environment with respect to landmine odor signatures, as well asmine detection dog operational perspectives for this specific detection task. The aim is to highlight that giventhe limited knowledge on the subject, there is a research gap that needs to be attended in order to efficiently es-tablish optimal operating conditions for the reliable performance ofmine detection dogs in Colombian deminingfield applications.

© 2015 The Chartered Society of Forensic Sciences. Published by Elsevier Ireland Ltd. All rights reserved.

Keywords:DeminingColombiaAntipersonnel minesMine detection dogsLandmine odor signatures

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 01.1. Colombia’s problem with antipersonnel landmines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 01.2. Description and design of antipersonnel mines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

2. Rapid overview of landmine detection methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02.1. Landmine detection process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 02.2. Current technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

3. Landmines and Canine Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03.1. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03.2. Canine landmine detection-the Colombian challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03.3. Landmine odor signatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03.4. Environmental parameters for landmine detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03.5. Chemical emission of landmine vapors in soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

nces. Published by Elsevier Ireland Ltd. All rights reserved.

dríguez, Demining Dogs in Colombia – A Review of Operational Challenges, Chemical Perspectives,), http://dx.doi.org/10.1016/j.scijus.2016.03.002

http://dx.doi.org/10.1016/j.scijus.2016.03.002

mailto:[email protected]

Journal logo


http://www.sciencedirect.com/science/journal/13550306

www.elsevier.com/locate/scijus


2 P.A. Prada, M.C. Rodríguez / Science and Justice xxx (2015) xxx–xxx

1. Introduction

1.1. Colombia’s problem with antipersonnel landmines

Colombia’s problem with antipersonnel mines and explosive rem-nants of war (ERW) is the direct result of over half a century of an inter-nal armed conflict. According to the Colombian government’santipersonnel mine action program, the period ranging from 1990thru June 9, 2015 registered a total of 11,133 victims due to antiperson-nel mines and unexploded ordnance (UXO). From these, 38%were civil-ians and 62%members of the Armed Forces. Just in the first trimester of2015, there were a total of 73 victims. The extent of this landmine con-tamination issue lies in 31 of the 32 departments within the country.The five most affected departments include: Antioquia, Meta, Caquetá,Nariño, and Norte de Santander [1]. Numerous survey and/or clearancetactical operations have shown that this man-made contamination islargely due to improvised explosive devices (IEDs)which act like an an-tipersonnel mine. Colombia’s largest rebel group, the FARC, are believedto be directly responsible for layingmany of themines and other IEDs inColombian territory. Furthermore, the National Liberation Army (ELN)even though not part of the current peace dialogues, continues toplant new antipersonnel mines in the department of Antioquia,among others. Since 2000, the FARC has increased their implementationas part of their response to the internal conflict across the country, thusmaking Colombia the only Latin American country to have a rising con-tamination issue over the past decade [2]. These illegal armed groupstactically use antipersonnel mines and explosive remnants as a way todeny millions of civilians and government authorities the access toland and natural resources that otherwise would increase the country’swelfare. Furthermore, the use of landmines represent a cheap way toprevent government authorities’ access to illegal economies such ascoca plantations, drug transport routes, clandestine laboratories, andguerrilla camps. The use of antipersonnel mines in strategic locationsalso serves as a terrorist mechanism that restricts and ultimately dis-places local communities who are forced to abandon their lands, indig-enous and Afro-Colombian communities being the most affected [3].

In view of this problematic, Colombian military forces in coordina-tion with the Presidential Program for Integrated Action against Anti-personnel Mines currently coordinate the operational activities withregards to humanitarian demining in affected territories. Currently,the National Police of Colombia has also joined forces and has imple-mented different mechanisms to help with the operational deminingactivities. Typically, the operational planning of humanitarian deminingimparts two phases, technical and non-technical. The technical phase isconducted directly on the field with direct physical intervention andtechnical equipment for humanitarian demining. The non-technical

Fig. 1. Schematic of land

Please cite this article as: P.A. Prada, M.C. Rodríguez, Demining Dogs in Coand Practical Implications, Sci. Justice (2015), http://dx.doi.org/10.1016/j.

phase studies entail the collection of data and information analysis inregards to suspected areas of mine contamination to identify the typesand dimensions of danger perimeterswithout any physical intervention[4].

1.2. Description and design of antipersonnel mines

The use of antipersonnel mines dates back to World War II as ameans to hinder opposing soldiers from clearing antitank mines.These original designs weremade from hand grenades and simple elec-tric fuses. Modern designs have greatly developed since then and cannow deliver fatal blasts of lethal pellets which can reach a radius of upto 100 m [5]. The basic components of a landmine include the activa-tion/triggering mechanism, the detonator (sets off booster charge),the booster charge (can be attached to fuse, detonator, or be part ofmain charge), the main explosive charge (bulk body of mine) and thecasing that contains all of these components (see Fig. 1) [6].

Illegal armed groups in Colombia manufacture their own landmineswith various activation mechanisms that are triggered by the victimsthemselves. These explosive devices are carefully dug underground inseparate locations approximately 5meters apart and connected by elec-trical wiring that are not easily detected by traditional mine detectiongear. The triggeringmechanisms utilized in Colombian territory includesyringes, tripwires, clothes-pin, mouse traps, wedges, and remote con-trols. Furthermore, common explosives used in their manufacture areammonium nitrate, mixtures of ammonium nitrate with fuel oils(ANFO),mixtures of sawdustwithALANFO (R1), hydrogels, nitroglycer-in, trinitroluene (TNT), to name a few [7]. Colombian landmine inci-dents are further complicated by the inclusion of elements such asfeces, nails, glass, and plastic scrap which cause wound infection notreadily detected upon medical inspection (Fig. 2). The mine casingthemselves are hard to detect as they come in various shapes andsizes of plastic materials (PVC pipes, bottles), textiles, glass, or evenwooden boxes. The problem also lies in the amount of explosive mate-rial Colombian rebel groups employ. Conventional mines contain from30 to 520 g of explosivematerial, while landmines in Colombian territo-ry report 250 g – 4 kg and some have yieldedmore than 20 kg of explo-sive content [8].

Landmines can be generally classified as either blast or fragmenta-tion. Blast mines are characterized by a shallow burial, and whose trig-ger mechanism originates from the pressure coming from a victim asthe subject steps on the mine. The activation weight for a typical blastmine ranges from 5–24 lb, making children the most susceptible vic-tims. The activation of a buried landmine causes the affected object(typically victims’ lower extremity) to blast into fragments in an up-ward direction, which is most of the cases the major cause of personal

mine componentry

lombia – A Review of Operational Challenges, Chemical Perspectives,scijus.2016.03.002


Fig. 2. Examples of antipersonnel mines manufactured in Colombia A) typical syringe triggering mechanism B) nails within mine capsule C) seizure of antipersonnel mines

3P.A. Prada, M.C. Rodríguez / Science and Justice xxx (2015) xxx–xxx

injury resulting in amputation of extremities [5,9,10]. Blastmines usual-ly have a cylindrical shape, 2–4 inches in diameter, and a height be-tween 1.5-3.0 inches. To make their detection harder, their casing canbe plastic or wood (instead of metal), although they can containminutemetal parts such as the firing pin or spring/washer mechanism [11].Some non-metal landmines found in Colombian territory are activatedby a downward pressure on a plastic syringe which releases sulphuricacid onto a mixture of potassium-nitrate with sulfur and carbon. Thismix subsequently detonates the main explosive [7].

Fragmentation mines, as the name implies, disperse fragments radi-ally outward at high speed. These type of mines cause damage atdistances up to 100m [5]. Some fragmentationmines are buried under-ground but upon activation are propelled upward exploding a meterabove ground, sending lethal fragments in the surrounding radius.Other types of fragmentation mines are simply mounted on stakes inthe ground. Their activation is often guidedby tripwires, thus anymove-ment at a distance of up to 20 m can trigger the mine before actuallyreaching its exact location. Hence, for the purpose of an efficient and op-timal demining effort, the clearance of tripwire plays a crucial role [9].

2. Rapid overview of landmine detection methods

2.1. Landmine detection process

Even though landmines have been used since World War II, in gen-eral, the basic immediate response detection gear has not fluctuatedmuch. A key tool is the metal detector, a prodding tool (i.e. stainlesssteel probe, pointed stick), and a tripwire “feeler” [12]. Probing is anestablished procedure in any demining process. In recent years, im-provement in probes provide enhanced detection methods such asacoustic, electromagnetic, or even chemical. The deminer scans thesoil at an angle of typically 300 [13]. The demining process entails divid-ing the suspected mined area into grids (approximately 100 m2), split-ting the grids into lanes (approximately 1 m wide), and carefullyadvancing down each lane with a metal detector close to the ground.If the detector gives a signal, the demining team probes the target areato determine if amine is present. If it is confirmed, themine is detonatedby qualified demining responders [9]. However, as mentioned earlier,because of low content ofmetal material in recentlymanufactured plas-tic mines, metal detectors are sometimes rendered useless for this de-tection applicability. Thus, the demining process is a slow procedurethat involves a high risk for personnel as it can cause the accidental ac-tivation of antipersonnel mines. Hence, these factors hinder the rapiddemining and as a result the number of installed mines far exceedsthose detected and deactivated.

2.2. Current technology

Landmine detection technology in practical use today by affectedcountries worldwide mostly include metal detectors and mine detec-tion dogs. However, in addition to these established methods, research


and development efforts have resulted in a number of alternatives forthe detection of explosives and landmines [13–19]. Even though it isnot the intent of this review to cover each category in detail, fourmain areas that current technologies fall into include: trace/vapor ex-plosive detection, bulk explosive detection, mine casing detection, andinfrared/hyperspectral detection (Fig. 3) [8].

Explosive trace detection makes use of low concentrations of explo-sive compounds and residues in soil or in the boundary layer of air at thesoil surface. It has been recognized that all landmines leak with timesmall amounts of explosives that move in and onto the surroundingvegetation and ground surface [15]. Thus, landmine explosive odor sig-natures represent a key indicator for these vapor detection methodolo-gies. They can be separated into chemical and biological methods.Chemical methods including mass and ion mobility spectrometry,Raman scattering and even polymers designed to interact with micro-electronics upon chemical reaction of explosive molecule. Biologicaltrace detection involves the use of mammals, insects or even microor-ganisms to detect explosives, the most widely used being trainedmine detection dogs. Bulk explosive detection encompasses technolog-ical developments such as neutron analysis methods and nuclear quad-rupole resonance (NQR). However, with neutron analysis methods, theradiation dose that a human operator receives when in use is a majordrawback. There is also a great concern about the costs of both neutronand gamma ray sources. Neutron methods involve distinguishing theexplosives in landmines from surrounding soil by probing the soilwith neutrons and/or detecting returning neutrons. Thus, detection isachieved by monitoring differences in the intensity, energy, and otherfactors of the returning radiation [9]. NQR sends radio frequency pulsesthat excite nitrogen nuclei in the explosive hence inducing an electricpotential at the receiver coil. NQR detection is highly specific and yieldslow false alarm rates. For the particular case of Colombian landmineswhich are homemade using ammonium nitrate and fuel oil, NQR alsoshows a high promise for detection [20,21].

Mine casing detection of explosives include both electromagneticand mechanical methods. Common electromagnetic methods beingthe traditional metal detector, ground penetrating radar, X-ray back-scattering, and electrical impedance tomography. Due to the lowmetal content of Colombian landmines, metal detectors are not as use-ful, and X-ray backscattering is not field portable hence not useful incomplex topography. The most promising mine casing detection isground penetrating radar (GPR) as it emits radio waves into the earthand consequently analyzes the returning signal produced by reflectionsat the discontinuities of the dielectric constant [8]. Furthermore, thistechnique has been successfully applied to Colombian soils highlightinghow GPR can contribute in optimal detection of low and non-metallictargets [22]. Mechanical methods that fall under mine casing detectioninclude seismic/acoustic systems that introduce sounds or seismicwaves which are reflected when a landmine is present causing a vibra-tion. Other techniques include ultrasound and prodding, which is thetypical confirmatory method for demining purposes. A final categoryof detection capabilities includes infrared/hyperspectral systems.



Fig. 3. Developed explosive/landmine detection methods


These methods detect anomalous variations in electromagnetic radia-tion reflected or emitted by either surface mines or the soil/vegetationin their surroundings. This category includes infrared cameras, millime-ter wave radar, visible light, and laser induced breakdown spectroscopy[9].

3. Landmines and Canine Detection

3.1. Background

A well-established role for dogs in odor detection tasks is that oflandmines, improvised explosive devices, undetonated munitions, andother explosive materials that inherently yield to security risks to civil-ian and military populations [23–26]. Detection of these explosivethreats by canine teams and their ultimate removal by technical deploy-ments can reduce the military and civilian casualty rates in affectedcountries, thus reopening land for displaced citizens and farmerswhile concurrently contributing to a general national security level[27,28].

The introduction of mine detection dogs for field detection was ob-served during and after World War II, with an increased frequency ofuse in 1989 due to the first humanitarian mine clearance operation inAfghanistan [29]. Canine use for landmine detection beganwith very lit-tle research and deployment activities worldwide varied in their opera-tional structure from country to country. In 1999, a convention inLjubliana recognized the fact that a research gap existed in the optimalperformance and standards for the operational use of mine detectiondogs. This resulted in the creation of a United Nations approved interna-tional standards formine detection dogs, the IMAS 09.4 series. Thus, dif-ferent demining organizations worldwide can now implement internalstandards to ensure compliance with IMAS guidelines. These guidelines


are voluntary and thus each country approaches operational proceduresconsidering their own local needs [30]. The IMAS 09.4 guidelines are di-vided into five separate documents which cover a general guide for theuse of mine detection dogs, operational procedures, operational testingof mine detection dogs and handlers, remote explosive scent tracing(REST), and a guide to general medical and health care of dogs [31].

A crucial advantage of canines for landmine detection over technicalequipment is that they are not distracted by metal contamination andtheir inspection rate is much faster than a human. The two main rolesof a dog for demining purposes are to 1) delimit safe areas for manualdeminers and 2) provide a quality control check of areas cleared byother technical resources. If a dog detects a mine, the canine is removedfrom the area which is subsequently investigated by manual deminerswho proceed with the excavation and clearance. Hence, it can be saidthat canines are a detection tool to be used in combination with techni-cal equipment for optimal clearance operations [32].

3.2. Canine landmine detection-the Colombian challenge

The Colombian crisis with antipersonnel landmines highlights theneed for detection systems that are guided more by explosive vapor/trace detection due to the wide range of materials employed in theirmanufacture. As a result, canine detection is a vital tool in need of fur-ther development and crucial in any deployment activity in minedareas. Themajority of the affected territories in the country cannot sim-ply be called “mine-infected areas”, as the landmine distribution doesnot have a predictive or structured pattern and the quantities of minesis highly variable. This situation further complicates the demining activ-ities in Colombian territory. Even though employing canine teams to aidin the delineation of safe areas is a necessary tool for this humanitarianprocess to be achieved, it is important to understand some limiting




factors and challenges encountered in Colombian territory that directlyaffect the ultimate search capabilities of mine detection dog teams.

One of the first factors to consider is actual deployment to themine-contaminated areas. It is complicated for the canine team to reach theaffected zone in the middle of the forest via air transportation such ashelicopters, as these can be impacted by rebel groups. This obstacleleads to the deployment of canine teams on foot, which can be approx-imately from 2–6 hours before they reach the target zone. Thus, thequality of canine olfaction once it reaches the destination is greatly re-duced and impacted due to exhaustion as they begin the explosive de-tection task. Furthermore, in some of these sites, rebel groups havethe majority of authority over the land, placing at risk the canine teamor simply allowing the canine to reach the mined area with the aim ofdirect landmine activation. Besides the logistical obstacles to reach tar-get mined areas, the climatic conditions of the Colombian forest regionsmakes the terrain difficult to work in. Most of the sites are swamp-likeafter heavy downpours intrinsic to these regions, factor that goesagainst the canine search activity, given that the odor particles willnot be emitted in a regularmanner. There have been caseswhere the ac-tual improvised explosive device has been physically moved as a resultof sub-surface soil movements due to water currents. Once the waterhas dried and currents have stopped, the device starts to dry and hasthe danger of being detonated with simple direct contact with theground. Other complicating factors are health related due to endemicdiseases present in the region which place both canine and handler atrisk. For example, canine leishmaniasis is amajor global zoonosis poten-tially fatal to humans and dogs, being particularly endemic to SouthAmerican regions. The life cycle of this disease involves two hosts, a ver-tebrate (including rodents) and an insect (sand fly) [33,34]. Theswamped area regions are prone to the availability of these vector-host interactions, thereby adding a challenge in the operational setting.Given that much of the affected area lies in thick forest and mountain-ous regions of the Colombian territory, along with the fact that illegalrebel groups use these regions for illicit crop growing and combatareas, makes canine deployment even more difficult. Operationaldemining activities cannot pursue organized search perimeters asoutlined in the IMAS guidelines, which causes the canine to air scent

Fig. 4. Colombia National Police Mine Detection Dogs A) Operational traini


explosive particles instead of direct tracking. This search criteria causesincreased risks, as the erratic movement of the search from side to sidein an unorganized manner could trigger the activation of an explosivedevice through contact with tripwires or inadvertently pressing activa-tionmechanisms of buried devices. Even though the risks are greatly in-creased, to date, there has been no better procedure to approach thegiven search conditions.

The first explosive detection canines were trained in 1987 in BatonRouge, LA, where the first Belgian Malinois were also brought for usewithin the National Police. These explosive detection canines were notinitially assigned specifically for landmine detection, as they were de-ployed to urban areas. Gradually, these canine teams were focused tomore rural areas and in 2007 operational demining procedures startedwithin the National Police of Colombia in manual illicit crop eradicationzones (Fig. 4). When these missions began, the selected breeds for pro-visional use were Labradors, while Belgian Malinois were being pur-chased for subsequent use, as it was considered this breed was moreresilient to operational needs. Nevertheless, the type of Labrador usedin Colombia by theNational Police and themilitary forces is amix of na-tive breeds that has demonstrated high performance and climate adap-tation, which has led to the decrease use of Belgian Malinois for thisdetection task as originally envisioned. Currently, missions have gearedto approach the problem combining both rapid operational procedureswith humanitarian perspectives, identifying target territories and clear-ing land to restore public confidence working for a post-conflictresolution.

The Colombian National Police has played an essential role in thetraining of mine detection dogs, not only for the police forces but alsofor other inter-governmental agencies such as the Colombian Armyand Air Force. In the realm of mine detection applications, the trainingfor these canine teams takes approximately three (3)months, includinga basic explosive detection training regime. However, the introductionof novel odors such as those obtained from seized local landmines is fur-ther utilized in their odor recognition assessments. These may includeone (1) additional month per new odor. Although, the variety oflandmines manufactured by the rebel groups is constantly evolving, ata minimum a canine team ready for operational deployment must

ng field site B) successful detection of a landmine-contaminated area




have had training with a range of 4 to 6 types of explosive odor signa-tures from seized landmines and/or IED componentry.

Within the National Police force, these mine detection dogs havealso been employed for the dual purpose of helping with manualeradication missions. Manual illicit crop eradication efforts have beena direct response by the government to deal with the production of co-caine since 2006. Traditionally, this operational mission was handledthrough aerial spraying with glyphosate herbicide. Due to the toxicityexposure to inhabitants and to the environment, the Colombian govern-ment made the switch to a manual eradication alternative. Farmers areusually hired to “physically” root out the coca plants which has yieldedto a reduction of 48,189 hectares of sowing area in 2013 compared to102,000 hectares of planted coca plants in 2002. These “eradicatingfarmers” need to reach coca fields alongside police and military forces.Within the Colombian National Police, the Antinarcotics Division hasbeen in charge of such a deployment mission. Within this group, thereis currently 3 canine teams dedicated to aid in this manual eradicationoperation. For approximately 10 years, canine teams have supportedthis national mission by helping to clear the rural and vegetationpaths leading to the coca plantations. As an added challenge to the dif-ficult working terrain, these canine teams confront the risk of encoun-tering implanted IEDs on the periphery of these illicit coca growingregions. This caused team casualties as the canines had to be trainednot only on a focused ground odor-tracking but also on particle airscenting as the IEDs in the Colombian jungles are routinely implantedin bushes and/or plants well-above ground level. As a result, thesecombined “mine-detection/crop eradication” canines had to offset tra-ditional IMAS guidelines to circumvent the challenges in this new oper-ational perspective. Table 1 highlights some traditional training/operational factors to consider and contrasts the IMAS guidelines tothe Colombian operational perspectives.

3.3. Landmine odor signatures

Canine detection is a type of explosive vapor detection techniquewhich is based on the successful response of the dog to landmine odorsignatures emanating from the buried devices. Whether the caninemakes the odor recognition from a single chemical or a mixture ofchemicals from both the explosive and the mine casing itself is stillnot known with certainty. However, much of the evidence in explosivedetection indicates that the extremely low vapor pressures for many ofthe common explosives further complicates the direct detection ofthese compounds [35]. Thus, research has geared to the identificationof target explosive odor signatures with higher vapor pressures thatare minor components or impurities of the explosive mixtures whichthemselves may be used in the detection of the parent explosives and

Table 1MDD Operational Comparison: IMAS vs. Colombian National Police Procedures

Operational Procedures

Conventional IMAS Colombian Tra

Safety lanes Clearly marked safety lanes used to provideaccess to and around demining worksite

Only in trainin

Search patterns Search-lane or short-leash systems Search lane: Thperforming theFor training pu

Numbers of MDD used At least 2 canine teams for each search area At least 3 caninFor crop eradicplants per hect

Environmental Factors Wind, rain, snow, humidity, atmosphericpollution, vegetation

Sloping hills/mTraining Site: Tterrains such thThe site is an awell as giving tcrop growing sIn terms of winodor signature


can contribute to the available odor picture [36–38]. TNT is one of themain explosive sources used for the manufacturing of landmines.Vapor generated from solid TNT is predominantly 2,4- DNT and 1,3-DNB, which comprise a small fraction of the solid phase. Nevertheless,the vapor pressure of 2,4- DNT and 1,3 –DNB are approximately40–1000 times greater than the vapor pressure of TNT, thus makingtransport into soil much faster for these two odor signatures than theparent compound. Other studies have focused onmeasuring the sensingthresholds for mine detection dogs to previously detected odor signa-tures for nitroglycerin smokeless powder and Composition-4. Byconducting laboratory and field tests of soil contaminated samples, re-markable ultra-trace levels of detection were observed with the testedcanines, which are on par with the chemical residues derived from theburied landmines [39]. The release of these identified chemical odorsignatures is guided by different paths within the surrounding area ofthe landmine having direct implications in the ultimate success fordetection.

3.4. Environmental parameters for landmine detection

Factors that need to be considered for optimal canine use for thistask lies not only in appropriate dog selection and training, but also onenvironmental conditions that directly impact landmine odor signa-tures in terms of chemical emissions, distribution, transport and degra-dation in the soil. As a result, the diversity of terrain and environmentalconditions within an affected area is a crucial factor that needs to beconsidered (Fig. 5).

Sargisson et al. studied environmental determinants of landmine de-tection canines in Afghanistan. Some key points that the studyhighlighted were that detection success decreased with increasingmine depth, and that heavier mines (higher explosive content) weredetected more easily. While other weather variables such as tempera-ture,wind speed and relative humidity seemed to have no significant ef-fects on detection, the results showed that humidity was the mostimportant variable. High humidity resulted in lower detection rates inarid environments, except when canine searches were conducted inthe early morning. Furthermore, higher levels of vegetation also re-duced detection success. However, aromatic plants did not seem to in-terfere with a successful detection [40]. The effect of temperature onthe flux of vapors emitted by several types of mines has also beenapproached. A vapor flux model developed by Leggett and colleagues[41] show that emission rates increase about three-fold for each ten de-grees rise in temperature. Similarly, in cold climates, vapor emissionsare much lower. Under snow conditions, when ground temperaturesreach approximately 0°C, vapor fluxes are expected to be ~1/9 of their

ining perspective

g exercises are safety lines marked and implemented.

e search system employed for MDD at all times is working the dog off-leash whilesearch. MDD are never worked with the short-leash system.rposes, both long and short leash systems are conditioned.e teams for each search area.ation missions, search area is divided per hectares. There are approximately 12.500are. Every suspected illicit crop area has a minimum of 3 hectares.ountain ranges; changes in altitude, heavy foresthe Colombian National Police has a unique training site simulating coca-growingat these MDD can get their training in simulated conditions as target search areas.ctive research facility where real coca plants are grown to study the crop process ashe canines an opportunity to work with environmental conditions that mirror illicitites (i.e. mountains, river, vegetation)d factor: Crucial problem for MDD search as the constantly changing gusts can affects from IEDs hidden in plant/bushes as well as tree heights (1.5-3.0 meters)



Fig. 5. Principal environmental factors affecting landmine odor signature emission


values at 20°C. Therefore, surface contamination is affected by externalconditions such as temperature, wind, and moisture [42,43].

Since explosives are labile, they are biotransformed by indigenousmicroorganisms, and photo degradation by sunlight as they migrateinto the subsurface soil matrix inwhich they are buried [44]. Thus, a sig-nificant part of the emitted landmine odor signatures can also be attrib-uted to the temporal and spatial variability that is present in soils.Surface temperature of soil varies approximately 30°C over a 24-hourperiod, largely depending on location, soil type and weather. Thus,this temperature gradient can and does desorb trace explosives fromthe surface thus creating a detectable airborne vapor signal [45]. Datacollected from soils in distinctive settings (temperate, tropical, desert)shows that soil moisture varies widely and over distances of less than1 m [46,47]. This variability has direct implications for sensors (i.e. ca-nines) that are affected by the soil water content, as their performancemay be variable over short distances. A thorough understanding of soilproperties can thus improve performance and explain observed pat-terns for optimal landmine detection [48]. In the valleys of the Andes,Oriental Plains and Andean region of Colombia, there is a great diversityof soils, soil quality and pathways of soil degradation caused by largevariations in parent materials, climatic conditions, biodiversity and thephysiographic position of the land [49]. In Colombia, volcanic ash soilsalso make-up the soil diversity of the affected terrain, however, generalknowledge of tropical volcanic ash soil is incomplete [7]. Therefore,there exists an immediate need for in-depth evaluation of Colombiansoil properties and its direct relation to emitted landmine odor signa-tures in this particular geographic region and its effect on deployed ca-nine teams in contaminated areas. These implications suggest thatdetection success should consider regional environmental parametersin the management of dog searching activities.

3.5. Chemical emission of landmine vapors in soils

The efficacy of biological (or chemical) detection and their potentialusefulness for detecting buried landmines cannot be approached with-out understanding how landmine odor signatures are transported andemitted into the soil. The transfer of landmine chemicals to soil involvesboth leakage and permeation, commonly referred to as landmine flux.Leakage occurs through the opening(s) of the mine casing. Permeationoccurs by vapor diffusion through the thickness of the mine case


material. The composition of the casingmaterial makes a significant im-pact on the leakage into the soil, hard plastics (i.e. PVC) permeating lesslandmine signature chemicals thanmore flexible materials such as rub-ber. Materials such as steel do not allow permeation. Therefore, perme-ation describes the rate at which the explosive vapor passes through aparticular material. Factors affecting permeation include the type ofpolymer, the nature of the vapor and the environmental conditions ofthe surroundings. Higher temperatures, for example, yield to higherflux rates. Thus, both leakage and permeation direct the chemical emis-sion for buried landmines [50]. Colombian landmines aremanufacturedfrom an array of differentmaterials. Common types includewhat are la-belled as “box mines” made from PVC or wood or “sack mines” whichderive their name from the way they are camouflaged inside sacks,making them appear as rocks on the terrain. The large variety ofmines encountered yields to a high number of variables such as compo-sition and manufacturing processes, which to date have not been re-corded or instrumentally analyzed in Colombian territory. For optimaldetector development it is essential to incorporate the plastic bodiesand/or other casing materials into the sensitivity testing of the sensorbeing employed [51]. This gap in knowledge demands immediate eval-uations in order to determine landmine fluxes given the current threatsencountered in this particular region.

Another key factor influencing chemical emission of landmine odorsignatures is external surface contamination. Transport of vaporsthrough the surrounding soil occurs in both the liquid and vapor phasesby diffusion and advection processes. Liquid phase advection occursthrough the precipitation and evaporation of water content from thesoil, while gas phase advection occurs as a result of barometric pressurechanges. Partitioning among the phases is important for explosive com-ponents, which tends to concentrate the explosives on the soil phaseand in the liquid phase. Furthermore, biodegradation also imparts an ef-fect on the explosive odor signature being emitted [52–54]. Degradationrates for 2,4-DNT and TNT have been measured to be 40 and 100 days,respectively [55]. Fig. 6 represents a schematic of the surface conditionsthat affect the generated landmine odor signature.

In a study conducted by George et al. a TMA5, a large plastic antitankmine of former Yugoslavian manufacture, revealed that concentrationsof explosive residues in the soil directly underneath the mine were100–10,000 times greater than surface concentrations. It was hypothe-sized that precipitation enters the soil and then flows around the buriedmine eventually becoming stagnant under the mine surface. The waterthat is trapped underneath themine increases the flux rate through theplastic as compared to the rate on the sides and the top of the minewhich are more affected by air flux rates. The compiled results of over1000 soil samples indicate that the concentrations of explosive-relatedresidues present in soils surrounding buried mines vary tremendouslydepending on the specific type of mine. Furthermore, chemical signa-tures were most often found in a discontinuous perimeter around themine, with very little to no signature detection directly over the mine[56].

4. Conclusions

In this review we have described the grave crisis of antipersonnellandmines in Colombian territory. Given the available literature inlandmine detection, this review serves to highlight this LatinAmerican country as the subject of study, as much of the published re-ports focus on research in countries like Cambodia, Afghanistan,Croatia, and Angola, but notmuch emphasis on the American continent.The large array ofmaterials used in Colombian landmines for theirman-ufacture by insurgents presents a crucial detection challenge. Thislandmine crisis poses different aspects of action tasks in areas of hu-manitarian demining, operational procedures, victim assistance, andrapid response times. It is inconceivable that given such a problematicissue a single detection method will eradicate the problem altogether,but the advancement of canine detection capabilities can provide a



Fig. 6. Schematic of chemical emission and transport processes in landmine odor signatures


valuable counterpart to comparable mine detectionmethods. The over-arching objective of this review is to focus on existinggaps in research toapproach the landmine and unexploded ordnance problem in Colombiaand present the potential role canine detection teams can play if usedeffectively for humanitarian demining operational procedures. TheColombian landmine issue presents the need to use canines as a key de-tection source given the wide variety of mine casing materials utilized,which are not necessarily of metal content easily detected by other de-tectionmediums. If proper training aids are developed and utilized dur-ing routine training activities, canine teams can then be exposed totargeted odor sources specific to the affected regions. The limited un-derstanding of the chemical odor signatures of the mine compositionmaterials (besides the explosive content) is a drawback to current oper-ational needs. It is essential to know the different odor signature combi-nations being currently seized and encountered in order to directtraining exercises accordingly. Furthermore, due to the implications ofenvironmental factors on the ultimate chemical odor emission of buriedlandmines, it is essential that a thorough evaluation of soil types andconditions be performed in order to direct canine searching activities.An evaluation of Colombian soil properties can then further the under-standing of explosive vapor transport and fate in the different sub-surface conditions of the country’s topography.

It is still a long way to rid this Latin American country of this deadlycontamination affecting thousands of innocent victims daily as a resultof decades of internal war conflicts. Hence, the urgent need to re-evaluate current operational guidelines and adjust existing standardsto regional requirements for enhanced results in giving back inhabitantsa landmine-free territory in which to live in peace.

References

[1] Presidential Program for Comprehensive Action against Anti-Personnel Mines(PAICMA). Accessible at http://www.accioncontraminas.gov.co/Paginas/aicma.aspx, Retrieved 9 June 2015.

[2] A. Moloney, Caring for Colombia’s landmine survivors, Lancet 373 (2009)2013–2014.


[3] S. Naidoo, C. Pardo, C. Theim Jensen, Mine Action and Land Issues in Colombia, Ge-neva International Centre for Humanitarian Demining (GICHD), Geneva,Switzerland, 2013.

[4] C.A. Garcia, A.P. Quimbaya, Spatial processing of events with landmines in Colombia,Colombian Computing Congress (CCC) Institute of Electrical and Electronics Engi-neers, 2012.

[5] J.G. Ackenhusen, Q.A. Holmes, et al., Detection of Mines and Minefields, VeridianSystems Division, Ann Arbor, MI, 2001.

[6] R. Bello, Literature Review on Landmines and Detection Methods, Front. Sci. 3 (1)(2013) 27–42.

[7] J.M.H. Hendrickx, A. Molina, D. Diaz, et al., Humanitarian IED clearance in Colombia,SPIE Defense and Security Symposium, International Society for Optics and Photon-ics, 2008.

[8] L. Cardona, J. Jimenez, N. Vanegas, Landmine Detection Technologies to Face theDemining Problem in Antioquia, Dyna, 812014 115–125.

[9] J. MacDonald, J.R. Lockwood, et al. Alternatives for Landmine Detection. Ed. RandScience and Technology Policy Institute, Santa Monica, CA, 2003.

[10] S.I. Ivashov, V.V. Razevig, A.P. Sheyko, I.A. Vasilyev, A Review of the Remote SensingLaboratory’s Techniques for Humanitarian Demining, Int. Conf. Requirements andTechnologies for detection, Removal and Neutralization of Landmines and UXO,Brussels, Belgium, 2003.

[11] P. Horowitz, K. Case, et al., New Technological Approaches to HumanitarianDemining, MITRE, JSR-96-115McLean, VA, 1996.

[12] A. Carruthers, J. McFee, Landmine Detection: An Old Problem Requiring New Solu-tions, Can. Def. Q. 25 (4) (1996) 16–18.

[13] C.P. Gooneratne, S.C. Mukhopahyay, G. Sen Gupta, A Review of Sensing Technologiesfor Landmine Detection: Unmanned Vehicle Based Approach, 2nd InternationalConference on Autonomous Robots and Agents, Palmerston North, New Zealand,2004.

[14] M. Sato, The Path to Humanitarian Demining and International Cooperation, Sci.Technol. Trends Q. Rev. 19 (2006) 22–34.

[15] M.K. Habib, Controlled biological and biomimetic systems for landmine detection,Biosens. Bioelectron. 23 (2007) 1–18.

[16] E.M.A. Hussein, E.J. Waller, Landmine detection: the problem and the challenge,Appl. Radiat. Isot. 53 (2000) 557–563.

[17] C. Castiblanco, J. Rodriguez, I. Mondragon, C. Parra, J. Colorado, Air drones for explo-sive landmines detection, Advances in Intelligent Systems and Computing, 2 2014,pp. 107–114.

[18] S. Singh, Sensors- An effective approach for the detection of explosives, J. Hazard.Mater. 144 (2007) 15–28.

[19] J. Yinon, Field detection and monitoring of explosives, Trends Anal. Chem. 21 (2002)292–301.

[20] Z. Bielecki, J. Janucki, A. Kawalec, J. Mikolajczyk, N. Palka, et al., Sensors and systemsfor the detection of explosive devices - an overview, Metrol. Meas. Syst. XIX (2012)3–28.

[21] L. Cardona, J. Jimenez, N. Vanegas, Nuclear quadrupole resonance for explosive de-tection, Ingeniare Revista chilena de Ingenieria 23 (2015) 458–472.


http://www.accioncontraminas.gov.co/Paginas/aicma.aspx

http://www.accioncontraminas.gov.co/Paginas/aicma.aspx

http://refhub.elsevier.com/S1355-0306(16)30003-X/rf0005

















































[22] O. Lopera, N. Milisavljevic, Prediction of the effects of soil and target properties onthe antipersonnel landmine detection performance of ground-penetrating radar: AColombian case study, J. Appl. Geophys. 63 (2007) 13–23.

[23] K.G. Furton, L.J. Myers, The scientific foundation and efficacy of the use of canines aschemical detectors for explosives, Talanta 54 (2001) 487–500.

[24] I. Gazit, J. Terkel, Explosive detection by sniffer dogs following strenuous physical ac-tivity, Appl. Anim. Behav. Sci. 81 (2003) 149–161.

[25] B.M. Jones, Applied behavior analysis is ideal for the development of a landmine de-tection technology using animals, Behav. Anal. 34 (2011) 55–73.

[26] L. Lazarowski, D.C. Dorman, Explosives detection bymilitary working dogs: olfactorygeneralization from components to mixtures, Appl. Anim. Behav. 151 (2014) 84–93.

[27] A.A. Faust, C.J. de Ruiter, A. Ehlerding, J.E. McFee, E. Svinsas, A.D. van Rheenen, Obser-vations on military exploitation of explosives detection technologies, Proc. SPIE8017 (2011), http://dx.doi.org/10.1117/12.886391.

[28] H.G. Watts, The consequences for children of explosive remnants of war:Landmines, unexploded ordnance, improvised explosive devices, and clusterbombs, J. Pediatr. Rehabil. Med. 2 (2009) 217–227.

[29] D. Hayter, GICHD, Training dogs to detect tripwires, Mine Detection Dogs: trainingoperations and odour detection, GICHD, Geneva 2003, pp. 109–138.

[30] I. McLean, R. Sargisson, Detection of landmines by dogs: environmental and behav-ioural determinants, GICHD, Geneva, 2005.

[31] IMAS 09.40 Guide for the use of mine detection dogs, http://www.mineactionstandards.org/fileadmin/MAS/documents/imas-international-standards/english/series-09/IMAS-09-40-Ed2-Am3.pdf 2013.

[32] I. McLean, R. Sargisson, Mine detection dogs: operations, GICHD, Geneva, 2005.[33] L. Solano-Gallego, G. Miro, A. Koutinas, L. Cardoso, M.G. Pennisi, L. Ferrer, et al.,

LeishVet guidelines for the practical management of canine leishmaniosis, Parasit.Vectors 4 (2011) 1–16.

[34] F. Dantas-Torres, Canine Leishmaniosis in South America, Parasit. Vectors 2 (2009),http://dx.doi.org/10.1186/1756-3305-2-S1-S1.

[35] L. Senesac, T.G. Thundat, Nanosensors for trace explosive detection, Mater. Today 11(2008) 28–36.

[36] R.J. Harper, J.R. Almirall, K.G. Furton, Identification of dominant odor chemicals em-anating from explosives for use in developing optimal training aid combinationsand mimics for canine detection, Talanta 67 (2005) 313–327.

[37] J.M. Johnston, M. Williams, L.P. Waggoner, C.C. Edge, R.E. Dugan, et al., Canine detec-tion odor signatures for mine-related explosives, Proc. PIE 3392 (1998) 490–501.

[38] C.C. Edge, J. Gibb, R. Dugan, Comparative study of the vapor analytes of trinitrotolu-ene (TNT), Proc. SPIE 3575 (1998) 282–290.

[39] J.M. Phelan, J.L. Barnett, Chemical sensing thresholds for mine detection dogs, Proc.SPIE 4742 (2002) 532–543.

[40] R.J. Sargisson, I.G. McLean, J. Brown, H. Bach, Environmental determinants oflandmine detection by dogs: Findings from a large-scale study in Afghanistan, J.ERW Mine Action 16 (2) (2012) 74–81.


[41] D.C. Leggett, J.H. Cragin, T.F. Jenkins, T. Ranney, Release of explosive-related vaporsfrom landmines. U.S. Army Cold Regions Research and Engineering Laboratory,Technical Report ERDC/CRREL TR-01-6, 2001.

[42] D.C. Leggett, T.F. Jenkins, A.W. Hogan, T. Ranney, P.H. Miyares, External contamina-tion of landmines by organic nitro-compounds, U.S. Army Cold Regions Researchand Engineering Laboratory, Technical Report ERDC/CRREL-TR-00-2, 2000.

[43] T.F. Jenkins, M.E. Walsh, P.H. Miyares, J.A. Kopczynski, T.A. Ranney, V. George, J.C.Pennington, T.E. Berry, Analysis of explosives-related signature chemicals in soilsamples collected near buried land mines, U.S. Army Engineer Research and Devel-opment Center, Technical Report ERDC TR-00-5, 2000.

[44] A. Halasz, C. Groom, E. Zhou, L. Paquet, C. Beaulieu, S. Deschamps, et al., Detection ofexplosives and their degradation products in soil environments, 9632002 411–418.

[45] G.S. Settles, D.A. Kester, Aerodynamic sampling for landmine trace detection, Proc.SPIE 4394 (2001) 890–898.

[46] T.W.J. Bauters, T.S. Steenhuis, D.A. DiCarlo, J.L. Nieber, L.W. Dekker, C.J. Ritsema, J.Y.Parlange, R. Haverkamp, Physics of water repellent soils, J. Hydrol. 231 (2000)233–243.

[47] J.M.H. Hendrickx, B. Borchers, J.Woolslayer, L.W. Dekker, C. Ritsema, S. Paton, Spatialvariability of dielectric properties in field soils, Proc. SPIE (2001) 398–408.

[48] R.L. VanDam, B. Borchers, J.M.H. Hendrickx, Strength of landmine signatures underdifferent soil conditions: implications for sensor fusion, Int. J. Syst. Sci. 36 (2005)1–15.

[49] A. Garcia-Ocampo, Fertility and soil productivity of Colombian soils under differentsoil management practices and several crops, Arch. Agron. Soil Sci. 58 (2012)S55–S65.

[50] J.M. Phelan, S.W. Webb, Chemical Sensing for buried landmines- fundamental pro-cesses influencing trace chemical detection, Sand Report SAND2002-0909, SandiaNational Laboratories, 2002.

[51] Y. Das, Study of the influence of the plastic casing on the electromagnetic inductionresponse of a buried landmine, Proc. SPIE 6953 (2008), http://dx.doi.org/10.1117/12.776699.

[52] S.W. Webb, J.M. Phelan, Effect of Diurnal and Seasonal Weather Variations on theChemical Signatures from Buried Landmines/UXO, Proc. SPIE 4038 (2000) 474–488.

[53] S. Desilets, L.V. Haley, U. Thekkadath, Trace explosives detection for findinglandmines, Proc. SPIE 3392 (1998) 441–452.

[54] J.S. Lovell, P.D. French, Active sampling technique to enhance chemical signature ofburied explosives, Proc. SPIE 5415 (2004) 494–501.

[55] J.C. Oxley, J.L. Smith, E. Resende, E. Pearce, Quantification and Aging of Post-blast res-idue of TNT landmines, J. Forensic Sci. 48 (4) (2003) 742–753.

[56] V. George, T.F. Jenkins, J.M. Phelan, D.C. Leggett, J. Oxley, S.W. Webb, et al., Progresson determining the vapor signature of a buried landmine, Proc. SPIE 4038 (2000)590–601.













http://dx.doi.org/10.1117/12.886391








http://www.mineactionstandards.org/fileadmin/MAS/documents/imas-international-standards/english/series-09/IMAS-09-40-Ed2-Am3.pdf







http://dx.doi.org/10.1186/1756-3305-2-S1-S1











































http://dx.doi.org/10.1117/12.776699

http://dx.doi.org/10.1117/12.776699













science and justice - policia

Documents