Forensics and Geospatial Technologies: tools to understand the forced disappearance of 42 people.
- Alejandro Dever
- Jan 25, 2022
- 3 min read
Updated: Feb 8, 2022
By. Alejandro Dever
Image processing and maping software: Erdas IMAGINE, ArcMAP, Autocad
In the 1980's and early 1990's Colombia was immersed in a low intensity civil war between Right wing paramilitary groups, the government forces and Radical Left Wing Guerrillas Paramilitary groups would often use forced disappearance as a way to further their agendas, exercising power over local populations and cementing the control of landowners over their territories. The Pueblo Bello case (named after a small town in the province of Cordoba, north of Colombia), is the story of 42 people who were murdered and burried in mass somewhere nearby the town. In 2004 I was approached by the Equitas forensics group and worked as the research teams geospatial technology consultant. For the next 3 years and taking into consideration the known behaviors of the paramilitary group dominant in the area, the phisiography of the region and my knowledge of systematics I proceded as follows.
Identify the Problem: Where where they murdered, where are the remains?
Formulate a Hypothesis: The operation of Paramilitary groups is related to the large cattle ranches that operate in the province of Cordoba. Murder sites would be accessible via private roads inside the haciendas, and not easily visible (avoided cattle grazing areas because dry grass would produce an unussually green patch of tall grass as bodies decomposed). Likely grave sites are in areas easily excavated and camuflaged. This led to a process of elimination of areas with incompatible land use, and the beginings of a working hypothesis and a model on the paramilitary behavioral pattern. Landsat imagery was used to classify land use and single out areas that would be ideal places to hide illegal mass graves. The areas with Tall humid forest, manioc and banana plantations were identified as serving all the conditions for this purpose.
In the mid 1990's, using informants, the police retreived the remains of 30 people and burried them in a mass grave in the towns cemetery. The forensic analysis was minimal and the remains of 18 people remained unaccounted for. Furthermore, changes in the course of the river over suggested that the remains were eroded and washed away at some point during the mid 1990's and early 2000's. Historical analysis of remote sensor imagery, (see Images A,B,C,D below), was used to map the changes in the Sinú river and use this data to aid in the field survey, and recovery program. It was determined that the river had changed course and eroded two areas of interest that had all the characteristics of ideal illegal grave sites.
A semi systematic pedestrian survey moved in to identify, and verify the hypothesis. A high resolution Quickbird image, and GPS was used to map sample sites, excavations and results. Previously flooded ares could be easily eliminated from the search area, small excavations verified that soil dominated by undisturbed silt, imagery supports the fact that these areas were flooded at the time of the murders. The main ares were identified Sites 1, 2,3, 5 and 6. Site 1 and point A showed the confluence of multiple private roads, exposed soil, disturbed soil and patches of mid sized forest and structures that may have been used as places used for criminal activity, perhaps torture, subsequent studies outside the scope of this specific survey showed to be mass grave sites inside the boundaries of this image. Site 2 and 5 had patches of Mid sized forest and planation, private roads and would be at the perifery of large cattle ranches. Site 6 was singled out in the the innitial plan as a preferred candidate and this was confirmed by the excavation of the remains of the people that had evidence of being murdered and tortured. Site 4, was tested but it was not considered a likely candidate.
Quickbird composite image: The Area of Interest measures 6 square kilometers. Site 5: Contained the presumed site of the murders, torture and material evidence of the murders.
Conclusion: The sectors (Playa Caudillo and Southern sector, see below) that were hypothesized to have ideal conditions for mass graves did have higher than expected positive results. The model based on the known behavior of the perpetrator groups in relation environmental factor and land use showed to have a high likelyhood of success in finding positive sites, but also equally important negative sites.
Comments