A GIS Model for Minefield Area Prediction: The Minefield Likelihood Procedure

• Main Author: Chamberlayne, Edward Pye
• Other Authors: Civil Engineering
• Format: Others
• Language: en
• Published: Virginia Tech 2014
• Subjects: likelihood; MCS-Eng; ABCS; NRMM; IMSMA; prediction; minefield; GIS; DTSS
• Online Access: http://hdl.handle.net/10919/36183; http://scholar.lib.vt.edu/theses/available/etd-12162002-135105/

Existing minefields left over from previous conflicts pose a grave threat to humanitarian relief operations, domestic everyday life, and future military operations. The remaining minefields in Afghanistan, from the decade long war with the Soviet Union, are just one example of this global problem. The purpose of this research is to develop a methodology that will predict areas where minefields are the most likely to exist through use of a GIS model. The concept is to combine geospatial data layers to produce a scored raster output of the most likely regions where minefields may exist. It is a "site suitability analysis" for minefield existence. <p> The GIS model uses elevation and slope data, observer and defensive position locations, hydrographic features, transportation features, and trafficability estimates to form a minefield prediction surface. Through use of the NATO Reference Mobility Model (NRMMII) and the Digital Topographic Support System (DTSS), trafficability estimates are generated for specific vehicles under specific terrain and weather conditions in specific areas of interest. <p> The model could be used to create prioritized maps for minefield detection sensors, demining teams, or for avoidance. These maps could define the "high payoff" search areas for remote sensors, such as ASTAMIDS, and positively identify minefields. These maps could also be used by humanitarian relief agencies for consideration when planning movement into areas that may contain minefields. The analysis includes a model calibration and sensitivity analysis procedure and compares the model output to known training minefield locations taken from two US Army training centers. The resultant Minefield Likelihood Surface has a 91% accuracy rate when compared to known training minefield data. === Master of Science