Mexico鈥檚 missing persons crisis has intensified over the past 20 years, with putting the number of people unaccounted for at more than 100,000. Now, the most extensive model ever made for predicting the locations of clandestine graves in Mexico could help track down the remains of those who have disappeared.
is part of a team of researchers at , a geospatial research centre funded by the Mexican government, that has been experimenting with using satellite imagery to identify the footprints left by covert graves.
As scores of people went missing over the past few decades, the government and groups of relatives of those who had disappeared began to ask CentroGeo for help, says Alegre Mondrag贸n. 鈥淪o, we started looking for variables that could help us answer these questions.鈥�
Advertisement
They gathered geographic data from known covert grave sites and combined it with information about the victims and perpetrators. For example, they focused on migratory routes because the people who use them are a particularly vulnerable group to disappearing in Mexico. They also used information about drug cartels鈥� , which are known to be behind many of these disappearances, .
Their model combined 30 variables, including road networks, bodies of water and urban settlements. Then they created maps of 10 of the states most affected by disappearances, highlighting areas where covert graves may be more likely to be discovered. For example, they found there was more potential for finding clandestine graves in areas with low population density and that were less than 46 minutes of travel time from roads in urban areas. In fact, 85 per cent of graves that have already been found were within 850 metres of a road. 鈥淧eople are innately lazy. They will do the least amount of work possible to get what they need done鈥�, and this applies to selecting a covert grave site, says at Deakin University in Australia, who wasn鈥檛 involved in this research.
But there were distinct differences across the 10 states. For most of them, clandestine graves were more likely when visibility was low 鈥� that is, when there were more obstacles, such as trees, blocking the view of the grave site. The exceptions were the three states of Michoac谩n, Jalisco and Guerrero, where this pattern was inverted.
Alegre Mondrag贸n suggests that this may be due to a high level of organised crime in these states. The level of control they have over a territory is such that they aren鈥檛 concerned about being found, she says.
Berezowski says that the scale of conflict in a region may also play a part. Larger conflicts may not follow the patterns of smaller ones. 鈥淥ften in these situations of mass conflict, the motive for burying [someone] in a mass grave is just to get rid of the evidence, so the variables that make sense for finding individual burials don鈥檛 necessarily work for mass burials.鈥�
Despite these challenges, these techniques are beginning to show modest success in Mexico. An earlier model designed only for the state of Baja California assisted a search brigade last year that found .
鈥淭his was very encouraging for the implementation of the model, because it was the first time it was applied in the field and it gave positive results,鈥� says at Data C铆vica, a human rights organisation in Mexico that collaborated with CentroGeo in the development of the earlier model.
Moving forward, Franco Boscan believes transparency is key. Mexico has no official record of clandestine graves, so 鈥渢o be able to run the model, first we have to generate that data鈥�. She and her team at Data C铆vica make frequent requests to prosecutors鈥� offices in Mexico to get the coordinates of discovered covert graves, she says. 鈥淭hat is the main technical challenge of the process.鈥�
Forensic Science International
