Chapter 3 describes the methods used to analyse whether political violence is more clustered near transport infrastructure, how the intensity of transport-related violence has changed over time, and which regions are most affected by violence targeting infrastructure in North and West Africa. The spatial and temporal relationships between violence and transport infrastructure are examined using road data from the Global Roads Inventory Project (GRIP) and conflict data from the Armed Conflict and Location and Event Data (ACLED) project, covering the period from 2000 to 2024. Violent events are mapped according to their distance to several categories of roads over time. This quantitative analysis is complemented by a textual analysis of road-related incidents, documenting local dynamics of violence along road corridors and illustrated with selected cases of violent attacks targeting transport infrastructure in the region. The report also uses the Spatial Conflict Dynamics indicator (SCDi) developed by the Sahel and West Africa Club (OECD/SWAC), to identify major clusters of violence in the region.
Roads and Conflicts in North and West Africa
Abstract
Competition for control of roads, rail lines, and other transport networks among state and non-state actors drives diverse patterns of violence, fundamentally shaping the conflict landscape in North and West Africa.
Violence near infrastructure is evolving. Over the past two decades, the intensity and proximity of violence to transport routes have shifted, highlighting how conflict actors adapt their strategies in response to infrastructure.
Certain transport corridors, such as those in the Central Sahel and Lake Chad basin, experience recurrent attacks, illustrating how localised acts of violence are part of a broader, interconnected conflict dynamic.
Combining spatial analysis with case studies of transport-related violence offers actionable knowledge to guide infrastructure protection and conflict mitigation strategies.
As state and non-state actors compete for the control of roads, railways, and other means of transportation, they produce various patterns of violence that shape the North and West African conflict landscape. Assessing how this unique geography relates to the transport infrastructure is key to understand the spatial and temporal evolution of armed conflicts in the region. This report documents these relationships, by examining whether conflicts are predominantly clustered near transport infrastructure, how the intensity of transport-related violence has changed over time, and which regions are the most affected by violence targeting infrastructure (Table 3.1).
The objective of the first question is to determine whether violent events observed in North and West Africa from 2000 to 2024 tend to be primarily situated near transport infrastructure. In other words, are regions close to transport infrastructure more prone to violence than others? The report then investigates whether political violence has become more concentrated around transport infrastructure over the past 24 years. Finally, the third question examines which portions of the transport system are the most affected by violence. Are some transport corridors more violent than others?
The methodological approaches to address these questions are similar to those used by OECD/SWAC (2022[1]; 2023[2]) to study the role of borders and cities in North and West Africa. These approaches combine quantitative analysis of conflict data with qualitative studies of specific conflict regions and events.
To determine whether political violence is clustered near transport infrastructure, the report assesses the relative number of violent events according to their geographical distance to different categories of roads. If transport infrastructure is both a facilitator and a magnet of violence, as described in Chapter 2, then the highest concentration of violence should be observed near roads and decline significantly with distance from them.
A similar approach is used to examine whether political violence has become increasingly clustered near infrastructure, using the changing proportion of violent events located at certain distances from different categories of roads. An increasing concentration of violence near roads could be the sign that government forces are trying to reconquer a rebellious region, or that militants are increasingly targeting convoys on certain road corridors. Conversely, violence could move away from transport infrastructure if militants increasingly expand their activities in rural regions.
Finally, the report identifies several transport corridors along which political violence is particularly high. A textual analysis of road-related incidents provides a better understanding of the local dynamics of violence along those corridors. This quantitative analysis is illustrated with a selection of violent attacks that have targeted transport infrastructure in the region.
The report adopts a multiscalar approach to examine the relationship between transport infrastructure and violence in North and West Africa. At the regional level, the study covers 21 countries: Algeria, Benin, Burkina Faso, Cameroon, Chad, Côte d’Ivoire, Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Libya, Mali, Mauritania, Morocco, Niger, Nigeria, Senegal, Sierra Leone, Togo, and Tunisia (Map 3.1). This regional focus is complemented with a study of two transnational regions in which transport infrastructure have been particularly affected by violent attacks since the early 2010s: Mali and the Central Sahel, and Nigeria and the Lake Chad region. Compared with earlier studies (OECD/SWAC, 2022[1]; 2023[2]), the Central Sahel study area has been extended to the south to account for the expansion of jihadist groups towards the Gulf of Guinea (Box 3.1).
The current expansion of violence in West Africa is often described as a shift from “Sahelian” to “coastal” countries (KAS, 2022[3]). This categorisation must be treated with caution, given the strong geographical and historical links between the Sahel and the Gulf of Guinea coast.
While the Sahel is technically defined as a transition zone between the Sahara and the savannah, receiving between 200 and 600 millimetres of rain annually on average, rainfall variability makes any effort to use isohyets to define its boundaries challenging. All “Sahelian countries” transcend bioclimatic zones. A significant portion of Mauritania, Mali, Niger, and Chad is located within the Saharan domain, which receives less than 200 millimetres annual precipitation, while the south of Burkina Faso, Mali, and Senegal are typical of the Sudanese domain, characterised by precipitation between 600 and 1 300 millimetres. For this reason, the Sahel should be seen as “a space of circulation in which uncertainty has historically been overcome by mobility” (Walther and Retaillé, 2021, p. 15[4]), rather than a bioclimatic zone or a series of countries.
The same is true for “coastal countries”, which the literature on jihadism typically identifies as Benin, Côte d’Ivoire, Ghana, and Togo (Clark and Zenn, 2023[5]; Financial Times, 2024[6]), although there are concerns about jihadist encroachment on Senegal, renewed jihadist activity in Mauritania, and new forms of jihadist expansion in Nigeria. A significant portion of such “coastal countries” is located in regions that are climatically more similar to the Sahel than to the Guinean domain, which receives 1 300 millimetres of rain per year on average.
The Sahel has also long been connected to the Gulf of Guinea economically, socially, and religiously. The north of Benin, Côte d’Ivoire, Ghana, and Togo is punctuated with more than 20 large border markets, such as Malanville, Tingréla, Bawku, and Cinkansé, which facilitate the circulation of agricultural and manufactured goods across climatic zones (OECD/SWAC, 2017[7]). In these regions located more than 600 kilometres north of the Gulf of Guinea, ethnic ties with the Sahel are extremely strong, as evidenced by seasonal and conjunctural movements of population. Several vehicular languages have developed throughout the region thanks to the influence of precolonial polities, trade networks, migration, or transnational pastoral groups. The Sahel and the north of coastal countries also share a common religious heritage, marked by the predominance of Sunni Islam and past episodes of jihad that have transcended climatic zones since the early 18th century (Miles, 2018[8]).
Reliable data on the evolution of transport infrastructure in Africa is difficult to find. The categorisation of roads, temporal availability, and spatial coverage used by existing datasets greatly varies (Table 3.2, Meijer et al. (2018[9]), OpenStreetMap contributors (2024[10]), CIESIN and ITOS (2013[11]), making international comparisons extremely challenging. In some African countries, very limited information is available on the types and conditions of roads, two key factors that influence the average speeds reached by users of the transport infrastructure. Currently, no road dataset provides up-to-date information on the geographical extent and condition of the road network over the past 20 years.
Against this background, this study uses the Global Roads Inventory Project (GRIP) dataset, a comprehensive and harmonised global dataset suitable for cross-national studies on the African continent (Meijer et al., 2018[9]). GRIP integrates urban, national, and supra-national road data from various sources and addresses discontinuities in road classifications within and between countries. GRIP classifies roads into only six categories, based on the United Nations Logistics Cluster classification, compared with eight categories for the Global Roads Open Access Data Set (gROADS) and 26 for OpenStreetMap (OSM). In a region where most roads are unpaved, this simplified categorisation facilitates the analysis of the distribution of violence. GRIP aggregates road data from 2000 to 2018, which significantly limits our ability to detect changes in the network over time. The fact that most of the road data were added after 2010 only partially compensates for this shortcoming.
The GRIP data is available both as a vector and raster datasets of road length and density at a 5- arcminutes resolution. This report uses the GRIP spatial vector dataset for the African continent, specifically for the 21 North and West African countries covered by the study. Eight different data sources were incorporated into the GRIP for these countries, with most using two or three data sources, ranging from OpenStreetMap to the United Nations’ International Steering Committee for Global Mapping. The only exceptions were Gambia, Guinea-Bissau, and Liberia which had one source each (Vector Map of the World, Level 1).
The total road length for all countries in the GRIP dataset is just under 1 million kilometres, classified into one of five specified categories: highways (2 031 km), primary roads (58 155 km), secondary roads (98 295 km), tertiary roads (765 223 km), and local roads (74 568 km). Road length varies widely among the countries, ranging from a minimum of 4 769 km in Gambia to a maximum of 117 932 km in Libya (Table 3.3). For analysis against the locations of conflict event data, the GRIP dataset is projected using a conic equal area projection for Africa.
Other data sources on roads across the region were evaluated and found unsuitable for this project based on factors such as locational accuracy, road network completeness, cross-national consistency in attribute information, or temporal coverage. For example, OpenStreetMap data lacked consistent spatial coverage beyond urban areas and had large gaps in its road attributes. The widely used gROADS dataset also had significant issues with mislocated roads and was outdated for studying the evolution of violence over the past two decades, as the data aggregates several years from 1980 to 2010.
Another data option evaluated was the commercial Michelin Maps, digitised from paper copies by Müller-Crepon (2023[12]) and Müller-Crepon, Hunziker and Cederman (2021[13]) between 1966 to 2017. The strength of the data set is the time-varying information and comparatively consistent coverage of the whole continent since the 1960s. In practice, however, such commercial road maps, drawn at a resolution of 1:4 000 000, are highly generalised for visual simplicity. As a result, the location of roads is less precise than in most high-resolution maps, with errors of around 5 kilometres expected when the data is georeferenced. This lack of spatial precision would have made calculating the distance between transport infrastructure and violent events problematic. Of the options available, the GRIP data presented the fewest overall issues.
Map 3.2 shows the distribution of highways, primary, and secondary roads across the region according to GRIP. It illustrates some of the major challenges faced by global road datasets:
First, the GRIP categorisation of roads does not always reflect actual differences between road types. For instance, very few roads are classified as “highways”, even in North Africa countries where the highway network is well developed. In Morocco, only the Casablanca-Rabat segment of the highway, completed in the late 1980s, is classified as a highway. Segments between Rabat and Tangier and between Casablanca and Marrakech, completed in the late 2000s, as well as recent extension of the highway network to Oujda, Beni Mellal, and Safi are missing. In Algeria, much of the recent extension of the highway network, spanning from the Moroccan to the Tunisian border, is also missing. This suggests that highways are often categorised as “primary roads” by GRIP.
Second, recent additions to the road network are not always visible, reflecting the fact that GRIP aggregates datasets from 2000 to 2018. Several important primary roads built in the last 20 years are still categorised as secondary or tertiary roads, such as the Nouakchott-Nouadhibou road inaugurated in 2004.
Third, spatial coverage varies greatly across countries. Senegal, Guinea, Liberia and Sierra Leone appear to have a very dense network of tertiary roads, while neighbouring countries relatively few. These differences most likely reflect a lack of regional harmonisation across several road datasets; some tertiary roads listed by GRIP should probably be reclassified as local roads, especially in heavily forested areas of Guinea, Liberia, and Sierra Leone. For these reasons, the report’s analysis does not focus heavily on the relationship between violence and road categories.
Note: to improve visibility, local and unspecified roads are not represented.
Source: GRIP data, (Meijer et al., 2018[9]). Cartography by the authors.
The report mobilises conflict data from the Armed Conflict Location & Event Data (ACLED) project to study the geography of political violence in the region. ACLED provides disaggregated, georeferenced information on violent events across Africa since 1997 (Raleigh et al., 2010[14]). Building on previous work addressing the geography of conflict in North and West Africa by the OECD/SWAC (2020[15]; 2021[16]; 2022[1]; 2023[2]), this study identifies eight categories of actors based on their communal, ethnic, or political goals and structure, and, where possible, on their “spatial dimension and relationships to communities” (ACLED, 2023, p. 25[17])
Actors can be formal organisations involved in violent activities, informal groups of people, or non-combatant categories (Table 3.4). Formal organisations include “state forces,” defined as collective actors that exercise de facto state sovereignty over a given territory, such as military and police forces from the region. Another type of formal organisation is “rebel groups”, whose political agenda is to overthrow or secede from a given state. Violent extremist organisations affiliated with the Islamic State or Al Qaeda, such as the Islamic State West Africa Province or the Group for Supporting Islam and Muslims (JNIM) are coded in this category. Splinter groups or factions that emerge from a rebel group are recorded as distinct actors. For example, Ansar Dine, the Macina Liberation Front, Al-Mourabitoun and the Saharan branch of Al Qaeda in the Islamic Maghreb are each coded individually before they merged to form JNIM in 2017. Informal groups are defined based on their social, ethnic or regional attributes, such as “Fulani Ethnic Militias”.
Notes: ACLED uses different names for state forces according to the regime they have served and the unit that participated in a violent event. These names were merged in the above table, except for Libya, where “military forces of Libya”, “military forces of Libya Haftar Faction” and “military forces of Libya Government of National Accord” are represented as separate entities. ACLED also identifies numerous categories of civilians, such as fishermen, farmers, health workers, teachers and women. These actors are regarded as “civilians” in this report and merged into a single category by country. These changes explain why the number of state and civilian actors is smaller than in previous studies (OECD/SWAC, 2020[15]; 2023[2]). Source: Authors, based on data from ACLED (2024[18]) available through 24 May 2024. Data is publicly available.
ACLED distinguishes between two types of militias, those defined by identity and those that pursue political objectives. “Identity militias” are a heterogeneous group of militants structured around ethnicity, religion, region, community and livelihood. They are often named after the locality or region where they operate, like the Borno Ethnic Militia in Nigeria. Self-defence groups such as the Volunteer for Defense of Homeland in Burkina Faso or Dan Na Ambassagou in Mali are coded as identity militias. “Political militias” are organisations whose goal is to influence and impact governance, security and policy in a given state through violent means, such as the Imghad Tuareg Self-Defense Group and Allies in Mali. Unlike rebel groups, political militias “are not seeking the removal of a national power, but are typically supported, armed by, or allied with a political elite and act towards a goal defined by these elites or larger political movements” (ACLED, 2023, p. 27[17])
Several categories of civilian actors are identified by ACLED. “Rioters” are unarmed individuals or groups engaged in disorganised violence against civilians, government forces or other armed groups during demonstrations, while “protesters” are unarmed demonstrators who engage in a public event without violence. Finally, “civilians” refer to the unarmed and unorganised victims of violent events identified by their country of origin. International organisations, foreign military forces, private security firms, and independent mercenaries engaged in violent events are coded as “external” and “other forces.” It is important to note that the ACLED database does not distinguish between the perpetrators and the victims of an attack, except for civilians who are, by definition, unarmed and cannot engage in political violence.
The analysis focuses on three types of violent events representative of armed conflict in the region: battles between armed groups and/or state forces, explosions and remote violence, and violence against civilians (Table 3.5). Because the report focuses on politically motivated violence, nonviolent actions such as strategic developments are not considered.
Battles are defined as “violent interactions between two politically organised armed groups at a particular time and location” (ACLED, 2023, p. 12[17]). They can occur between any state and non-state actors and involve at least two armed and organised actors. This category is subdivided into three sub-event types, depending on whether non-state actors or government forces overtake territory or whether there is no territorial change. Battles caused more than 124 000 fatalities in the region from January 1997 through June 2024, during just under 30 000 events. Armed clashes are by far the most represented type of battles, with more than 90% of fatalities. Battles are the deadliest type of violent event, with 4.2 people killed per event, rising to 5.7 victims per event when non-state actors retake territory.
Explosions and remote violence are defined as “incidents in which one side uses weapon types that, by their nature, are at range and widely destructive.” (ACLED, 2023, p. 16[17]). These events can be carried out using bombs, grenades, improvised explosive devices (IEDs), artillery fire or shelling, missile attacks, heavy machine gun fire, air or drone strikes, or chemical weapons. They account for 15% of the events and fatalities recorded in North and West Africa since 1997. Explosions and remote violence have killed more than 36 000 people since 1997 in nearly 11 000 incidents. They cause 3.4 victims per event on average, and 9.5 victims per event for suicide bombings, the deadliest form of sub-event recorded in the database over the period of observation.
Violence against civilians includes “violent events where an organized armed group inflicts violence upon unarmed non-combatants (...) The perpetrators of such acts include state forces and their affiliates, rebels, militias, and external/other forces” (ACLED, 2023, p. 18[17]). Violence against civilians accounts for 45% of the events and 35% of the fatalities recorded in North and West Africa since the late 1990s. The vast majority of the 85 600 civilian deaths and nearly 33 000 incidents observed in the region are caused by direct attacks. On average, 2.6 civilians are killed per violent event in the region.
Source: Authors, based on ACLED (2024[18]) data available through 30 June 2024. Data is publicly available.
ACLED also tracks protests and riots, but these represent a fundamentally different political process from armed conflict. For this reason, protests or riots are not included in the main analysis of the report. The resulting data includes 72 906 violent events and 246 031 fatalities from 1 January 1997 to 30 June 2024. Because the GRIP road data has only been available since 2000, the analysis of the relationship between transport infrastructure and conflict is limited to the period between 1 January 2000 and 30 June 2024, in which 233 624 people were killed in 70 095 incidents (Box 3.2).
In line with previous studies on the geography of armed conflict in the region, this report builds on disaggregated conflict data provided by ACLED (Raleigh et al., 2010[14]). ACLED is currently the most widely used, comprehensive, and up-to-date conflict dataset available to researchers and policy makers alike.
ACLED is particularly adapted to study emerging forms of political violence in Africa, due to its adaptive definition of violence, and inclusion of a wide range of state and non-state actors. Violent events reported by newspapers, social media, reports, and other media sources are manually added to the database, described, and coded consistently over time (Raleigh, Kishi and Linke, 2023[19]). The dataset is updated weekly based on over 13 600 sources in over 100 languages. ACLED has established numerous partnerships with local observatories and researchers around the world to minimize the number of false positive events, duplicates, and fake news that could artificially increase the number of violent events and deaths. These collaborations also ensure that most violent events observed in remote or poorly documented regions are reported.
While major progress has been achieved in producing geospatial data, there is no such thing as a perfect conflict dataset. All global datasets such as the Uppsala Conflict Data Project - Georeferenced Event Data (UCDP-GED) or the Integrated Crisis Early Warning System come with limitations. One of the limitations of ACLED is that the directionality of attacks cannot be determined: the data cannot be used to identify the perpetrator and the victim of a violent incident, except for violence against civilians.
Another limitation is that violent events come with a “spatial precision code” that determines to what extent the geographical coordinates of an incident are precisely known. If the source material mentions a small part of a region (such as “the area of Tanwalbougou” in Burkina Faso), the event is assigned to a town that represents that area (Tanwalbougou). These events represent 36% of all violent events recorded by ACLED in North and West Africa since 1997. If a larger region (such as “Zamfara State” in Nigeria) is mentioned instead of a location, the event is assigned to the closest natural location noted in reporting or the nearest provincial city (the state capital of Gusau) (ACLED, 2023[17]). Fortunately, these incidents represent only 1.5% of the total.
Finally, ACLED significantly improved its data collection process over time, which means that data collected in the late 1990s should be used with caution. More recently, military juntas have suspended several international media and threatened or imprisoned independent journalists in such countries as Burkina Faso, Mali and Niger. These authoritarian measures have turned certain regions into an “information desert”, which negatively affects the coverage of violent incidents. This is particularly true when government forces are involved in atrocities against civilians or attacked by rebel groups, in which cases the exact number of fatalities is often hard to determine. For this reason, this report mainly builds on the number of violent events, which is less subject to political manipulation than the number of fatalities.
A Spatial Conflict Dynamic indicator (SCDi) is used in this report to assess the changing geography of violence, over space and through time (Walther et al., 2023[20]). The SCDi combines two spatial properties of violence: the intensity of conflict across a region, and the distribution of conflict locations relative to each other. As presented in Figure 3.1, the patterning of violent events relative to each other is a different concern from conflict intensity: two regions may experience the same number of violent events but result in very different geographical patterns according to whether violence is rather spread or concentrated in a few places. The SCDi has been previously applied to all North and West Africa (OECD/SWAC, 2020[15]; 2022[1]; 2023[2]) using a uniform grid of 50 by 50 kilometres to subdivide the study area. It is calculated annually for each of these grid cells since 1997 and is made available to the public on the Mapping Territorial Transformations in Africa platform run by (OECD/SWAC, 2024[21]).
Source: OECD/SWAC (2020[15]), The Geography of Conflict in North and West Africa, https://dx.doi.org/10.1787/02181039-en.
Measuring the intensity of violence
Conflict intensity (CI) is the first spatial property measured by the SCDi. This metric identifies the total number of events in a given 50 by 50 km grid cell, for a given year. This number of events is then divided by the area of the cell, to allow for comparisons between zones. The resulting CI metric has a lower threshold of 0 if there are no events within a given zone during a given year and no upper threshold. As the CI metric increases from 0, it reflects an increasing spatial intensity of violence within a zone (Figure 3.2).
In addition to calculating the raw CI score for each zone, the SCDi also classifies a grid cell as higher or lower than an expected CI value. The expected CI value for North and West Africa is called the CI “generational mean”, the 20-year average conflict intensity between 1997 and 2016. The CI generational mean is 0.0017 events per square kilometre, or four events by cell. In this report, a zone is classified as high intensity if four or more events occur in a grid within a given year, and as low intensity otherwise.
Source: OECD/SWAC (2020[15]), The Geography of Conflict in North and West Africa, https://dx.doi.org/10.1787/02181039-en.
Measuring the concentration of violence
Conflict concentration (CC) is the second property measured by the SCDi. It refers to the distribution of conflict locations relative to each other. An average nearest neighbour (ANN) ratio is calculated to determine whether the patterns of violent events exhibit clustering or dispersion. The ANN ratio is calculated as the observed average distance among violent events in each zone, divided by the expected average distance obtained if the events had been distributed randomly (ESRI, 2019[22]). Like CI, the CC metric has a lower threshold of 0, with no conceptual upper threshold. ANN ratios smaller than 1 indicate clustering, while ratios greater than 1 indicate dispersion. For example, the distribution of events represented on the left-hand side of Figure 3.3 is clustered, as shown by its ratio of 0.5. A random distribution of the same number of events has a ratio of 1 while a dispersed distribution, represented on the right-hand side, has a ratio of 1.5 (Figure 3.3).
Source: OECD/SWAC (2020[15]), The Geography of Conflict in North and West Africa, https://dx.doi.org/10.1787/02181039-en.
Types and conflict life cycles
The SCDi identifies four types of conflict based on whether violent events are dispersed or clustered and are of high or low intensity (Figure 3.4). The first type are conflicts with an above-average intensity and a clustered distribution of violent events, suggesting that violence is intensifying locally. The second type are conflicts with a higher-than-average intensity and a dispersed distribution of events, indicating that the violence is accelerating. The third type applies to conflicts where there are fewer violent activities and most of them take place near each other, possibly indicating a decreasing range of violent groups. The fourth type, in which a lower-than-average intensity and a dispersed distribution of events are combined, suggests that a conflict is lingering, perhaps because opponents are highly mobile or unlikely to face protracted opposition.
Source: OECD/SWAC (2023[2]), Urbanisation and Conflicts in North and West Africa, https://doi.org/10.1787/3fc68183-en.
These four types are indicative of different stages in the overall lifecycle of a conflict (Walther et al., 2023[20]; Walther, Radil and Russell, 2024[23]). For example, when the locations of violent events are dispersed from one another, this commonly happened when violence either first emerged or receded in an area. Conversely, in areas where conflict persisted for multiple years without abating, violent events were often closely clustered together. Similarly, long-running conflict zones are often characterised by ‘hot spots’ of high levels of spatial intensity of conflict events while areas with lower intensity levels are typically on the periphery of such hot spots, indicating the potential of low intensity violence to be a hallmark of spatial spread of a conflict. Taken together, the four spatial types reveal insights about the dynamics of a typical conflict in North and West Africa. These are general trends, however, and not all sub-zones, places or localities will exhibit the same lifecycles between the SCDi types.
New local SCDi metrics
Since its launch in 2020, the SCDi has been regularly used to track key aspects of the geographies of violence in North and West Africa. The success of the SCDi as a conflict monitoring tool inspired the development of several new features (Walther, Radil and Russell, 2024[23]). One of the new features in the SCDi is the ability to track and identify the histories of conflicts in cell locations, which provides several advantages. For example, the SCDi now calculates the number of years that a conflict of any type occurred in each cell, called the SCDi years-in-conflict metric. Accordingly, using this metric to map cells that experienced violence each year but have no recent history of violence can highlight the spread of violence to previously peaceful areas.
Another new local feature of the SCDi is the ability to compare current CI and CC scores against the historical norms in those cells. The SCDi now calculates the local intensity and distribution of violence by averaging the annual metrics using the years-in-conflict metric. Comparing a cell’s current year score against the averaged local metric can indicate how violence is either intensifying/de-intensifying or clustering/dispersing. As with the years-in-conflict metric, mapping these trends can illustrate exactly where conditions are improving or destabilising, helping to identify where in the conflict lifecycle a cell may be (Chapter 4).
The relationships between transport infrastructure and armed conflicts are analysed using a geographic information system to associate the location of violent events with the existence of different categories of roads. Violent events recorded by ACLED each year are overlaid on the GRIP road dataset to examine how close politically motivated violence is to the road infrastructure at the regional level (Figure 3.5). The number of events observed in the region is then represented according to their relative distance from transport infrastructure. If, as one assumes, violence tends to cluster near roads, then the results should exhibit a clear distance-decay effect. The patterns of violence observed in the region are then broken down by year to examine whether transport infrastructure is becoming more targeted. This evolution is analysed by representing how the share of violent events located at various distances from roads varies over time. Finally, the data representing the relationships between transport infrastructure and violence are disaggregated by country to highlight intra-regional variations.
In the Central Sahel and Lake Chad region, the description of each violent event generated by ACLED is used to identify transport-related incidents based on several keywords, such as “roads”, “highway”, “convoys”, “ambush”, or “vehicle”, that are directly associated with transport infrastructure, the people who use it, or the vehicles used to move across the region. For example, ACLED notes that on 17 April 2024, a convoy of Malian Armed Forces (FAMa) forces struck an IED likely planted by JNIM militants on the road between Bandiagara and Bankass in the Mopti region (incident #MLI32366). These transport-related incidents are selected and mapped to provide a more detailed understanding of the strategies used by state and non-state actors to control mobility at the sub-regional level.
References
[18] ACLED (2024), Armed Conflict and Location & Event Data project (database), https://acleddata.com/.
[17] ACLED (2023), Armed Conflict and Location & Event Data Codebook, https://acleddata.com/knowledge-base/codebook/.
[11] CIESIN and ITOS (2013), Global Roads Open Access Data Set, Version 1.
[5] Clark, C. and J. Zenn (2023), Jihadist Terrorism Encroaching on Coastal West Africa.
[22] ESRI (2019), How Average Nearest Neighbor Works, https://pro.arcgis.com/en/pro-app/tool-reference/spatialstatistics/h-how-average-nearest-neighbor-distance-spatial-st.htm.
[6] Financial Times (2024), The Islamist Insurgents Threatening West Africa.
[3] KAS (2022), The Jihadist Threat in Northern Ghana and Togo: Stocktaking and Prospects for Containing the Expansion, Konrad Adenauer Stiftung.
[9] Meijer, J. et al. (2018), “Global patterns of current and future road infrastructure”, Environmental Research Letters, Vol. 13/6.
[12] Müller-Crepon, C. (2023), “State reach and development in Africa since the 1960s: New data and analysis”, Political Science Research and Methods, Vol. 11/3, pp. 663-672.
[13] Müller-Crepon, C., P. Hunziker and L. Cederman (2021), “Roads to rule, roads to rebel: Relational state capacity and conflict in Africa”, Journal of Conflict Resolution, Vol. 65/2-3, pp. 563-590.
[21] OECD/SWAC (2024), Mapping Territorial Transformations in Africa, https://mapping-africa-transformations.org.
[2] OECD/SWAC (2023), Urbanisation and Conflicts in North and West Africa, West African Studies, OECD Publishing, Paris, https://doi.org/10.1787/3fc68183-en.
[1] OECD/SWAC (2022), Borders and Conflicts in North and West Africa, West African Studies, OECD Publishing, Paris, https://doi.org/10.1787/6da6d21e-en.
[16] OECD/SWAC (2021), Conflict Networks in North and West Africa, West African Studies, OECD Publishing, Paris, https://doi.org/10.1787/896e3eca-en.
[15] OECD/SWAC (2020), The Geography of Conflict in North and West Africa, West African Studies, OECD Publishing, Paris, https://doi.org/10.1787/02181039-en.
[7] OECD/SWAC (2017), Cross-Border Co-Operation and Policy Networks in West Africa, West African Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264265875-en.
[10] OpenStreetMap contributors (2024), Planet dump, https://planet.openstreetmap.org.
[19] Raleigh, C., R. Kishi and A. Linke (2023), “Political instability patterns are obsucred by conflict dataset scope conditions, sources, and coding choices”, Humanities and Social Sciences Communications, Vol. 10/1, pp. 1-17.
[14] Raleigh, C. et al. (2010), “Introducing ACLED: An armed conflict location and event dataset”, Journal of Peace Research, Vol. 47/5, pp. 651-660.
[8] Walther, O. and W. Miles (eds.) (2018), Jihads and Borders. Social Networks and Spatial Patterns in Africa, Present, Past and Future, Routledge, London.
[23] Walther, O., S. Radil and D. Russell (2024), “The spatial conflict life cycle in Africa”, Annals of the American Association of Geographers, Vol. 15/2, pp. 456-477.
[20] Walther, O. et al. (2023), “Introducing the Spacial Conflict Dynamics indicator of political violence”, Terrorism and Political Violence, Vol. 35/3, pp. 533-552.
[4] Walther, O. and D. Retaillé (2021), Mapping the Sahelian Space, Oxford University Press, Oxford.