BOKO HARAM – A GEOSPATIAL ANALYSIS OF A TERRORIST GROUP USING SPATIAL DATA SCIENCE AND GEOSPATIAL INTELLIGENCE
Open Access
- Author:
- Frost, Joshua
- Graduate Program:
- Spatial Data Science
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- March 20, 2023
- Committee Members:
- Daniel Steiner, Thesis Advisor/Co-Advisor
Fritz Connor Kessler, Committee Member
Anthony Robinson, Program Head/Chair - Keywords:
- Boko Haram
Geospatial Intelligence
Spatial Data Science
Terrorism
Geospatial
GIS
G function
Africa
West
Central
Nigeria
Niger
Cameroon
Chad
Lake Chad
Distance-based functions
Descriptive spatial statistics
terrorist
Global Terrorism Database
START
RStudio
GeoDa
ArcGIS
proximity
Islamic State of West Africa
ISWA
al-Qaeda
Taliban
ISIS - Abstract:
- Boko Haram is a religious extremist terrorist organization that has mainly been active in the West and Central African countries of Nigeria, Cameroon, Chad, and Niger, starting primarily in 2009. The group’s insurgencies in those countries have increased and waned over the years, but its past and recent attacks on a range of targets, from private citizens to government, have and continue to threaten the stability of that African region, especially with fragment groups that have branched off of Boko Haram such as the Islamic State of West Africa. Due to the geographic scope of terrorism by Boko Haram at various scales (countries, states, towns), geospatial analysis of the attacks is necessary to help better understand the group and its incidents of terrorism. Geospatial analyses in this work include various spatial data science and geospatial intelligence methods for a descriptive and diagnostic analysis of Boko Haram attacks, compiled from the Global Terrorism Database (GTD), including proximity, point pattern cluster, spatial autocorrelation, and satellite image classification analyses. These spatial methods executed with various programs and tools, namely ArcGIS Pro, exploratory data analysis software package GeoDa, and R programming in RStudio, show the spatial patterns of terrorist attacks committed by Boko Haram as well as differences in these patterns of attacks based on certain incident characteristics, such as by year, target, attack, and weapon type. For example, the group perpetrated a large number of incidents on Private Citizens and Property targets in the northeastern Nigerian state of Borno, and the group also committed widespread attacks on Police targets throughout most of Nigeria and into western Niger. In addition, correlation analyses between five demographic-socioeconomic variables with incident counts and rates (incidents per 100,000 people) show statistically significant negative associations between incident counts/rates on the state level and literacy rates of people aged 6+ in Nigerian states. The other four variables, including adult literacy rate, population density, mean years of education, and GINI Index, are not statistically significantly correlated with states’ incident counts or rates. Furthermore, correlations of the number of incidents at state and local levels are considered with an environmental factor. Specifically, temporal fluctuations in the surface extent of Lake Chad’s southern pool, based on satellite imagery, are examined to determine if water variability in Lake Chad from 2009 to 2020, which could exacerbate social pressures, is associated with yearly fluctuations of terrorism in the Lake Chad area. These correlation analyses found no statistically significant associations between the change in surface water extent of the southern pool and the number of incidents per year in the state and local areas around Lake Chad. The findings presented in this paper can assist counterterrorism efforts against Boko Haram and other similar groups, specifically within the West and Central African region, which, with many combined efforts, can strengthen the stability of the affected region.