Off the grid: Identifying high-risk and unmapped areas in East Africa

Author

Cornelia Scholz

Source(s)
University of Copenhagen
“…Maps used to say, ‘There be dragons here’. Now they don't. But that don't mean the dragons aren't there.”

Lorne Malvo, Fargo 2014

‘There be dragons here’ used to be scribbled by map makers in olden days at the edge of maps, marking blind spots in the unknown areas, where nobody knew what was out there. Today our maps don't say that anymore, but this doesn't mean that gaps and blind spots no longer exist. Our trust in maps extends so far, that we tend to assume that if there is nothing showing on a map, there is nothing there. While in reality, these areas might be thriving places, home to families or providing livelihoods for whole communities. Currently a billion people are missing off the ‘World Map’ (1) and those who are left off, are at risk to be left out. Therefore, it is essential to fill into these data gaps and white spots, to render invisible lives visible on the maps of this world.

Most deaths caused by disasters occur in areas of conflict and fragile states. Fragility, conflict, and violence are major causes of humanitarian crises and account for 80% of all humanitarian needs. Disaster Risk Reduction (DRR) efforts often neglect conflict-affected states despite the fact that vulnerabilities of people living in those areas are exacerbated by the effects of conflict (2,3). To reach those most in need, the expansion of innovative humanitarian approaches like Forecast-based Financing (FbF) in conflict situation should be considered. FbF is a form of anticipatory action, an innovative approach of acting before a disaster hits to mitigate its impacts before the event. Expanding early action to new areas and new settings like conflict-affected regions to save lives and livelihoods, requires as much information as possible about the areas and the people living there. It is essential to know who lives where and how to plan humanitarian actions to reach those most in need. This makes the availability of mapping material a crucial element in the success of planning and conducting innovative humanitarian actions.

To work into this gap, this study identifies high-risk and unmapped areas in East Africa. The analysis focuses on areas that historically have been affected by floods, drought and high vulnerabilities exacerbated through a history of being affected by conflict. The aim is to identify and prioritize high-risk areas and guide volunteer mapping efforts to map the gaps in these areas. The created mapping material will enable the development of impact-based forecasting services for anticipatory action, by supporting the identification of what is more likely to be impacted and where, consequently supporting planning and conduction of early actions in those areas exposed to natural hazards, high vulnerability, and with a history of conflict.

The analysis behind the identification of high-risk areas consists of a geospatial analysis of areas in Sudan, South-Sudan and Ethiopia impacted by historic conflict events, exposed to flood and drought, and affected by high food insecurity. Identified high-risk areas which are not mapped in OpenStreetMap yet, are prioritized for mapping.

Analytical concept of identifying and prioritizing high-risk areas by overlapping Layer 1 (L1) unmapped areas, Layer 2 (L2) natural hazards, Layer 3 (L3) conflict and Layer 4 (L4) vulnerability. Source: own diagram.
Analytical concept of identifying and prioritizing high-risk areas by overlapping Layer 1 (L1) unmapped areas, Layer 2 (L2) natural hazards, Layer 3 (L3) conflict and Layer 4 (L4) vulnerability. Source: own diagram.

Each layer is built individually after carefully reviewing and selecting available data sets of each country. For the conflict layer, a density analysis of conflict locations between 2000-2021 is conducted to identify hotspots of historic conflict areas. The natural hazard layer includes geospatial information on the number of people exposed to floods and droughts per area. The vulnerability layer is country specific as available data sets and characteristics of vulnerabilities vary significantly among countries. For all three study countries, the vulnerability layer includes geospatial information on food insecurity indicators and locations of IDP and refugee camps. To identify where all these layers geographically overlay and add up with natural hazard exposure and existing vulnerabilities exacerbated by conflict, a weighted overlay analysis is conducted. The results show where layers geographically overlay, highlighting high-risk areas. This study focuses on the above-described geospatial analysis, but once the analysis is concluded, the power of crowdsourcing will be leveraged to create crucial maps of identified areas which are unmapped in OSM. This happens in three steps:

  1. The identified areas are uploaded as tasks to the MapSwipe App. In those tasks satellite imagery is viewed by volunteers, selecting those which show buildings and roads.
  2. The areas of selected imagery are then used in creating tasks for the Humanitarian OpenStreetMap Team (HOT) tasking manager. This allows volunteers from all over the world to contribute either individually or participate in ‘Mapathons’ of the Missing Maps Project to work on humanitarian mapping tasks.
  3. In a final step, field mapping is conducted, if people on the ground with local knowledge are available to add ground truth information in OSM. The combination of remote mapping and field mapping is the ideal in order to have as complete and accurate geospatial data as possible of an area. For example, a remote mapper could identify a building structure from areal imagery, while a local field mapper can then tag the structure as a critical facility of certain characteristics (e.g. construction materials, quality etc.).

Workflow. Mapping unmapped areas.

Through this process, maps of previously blind spots are created in OpenStreetMap. Since 2010 crowdsourced mapping has been mostly used for disaster response operations, but in recent years it has gained more traction in supporting disaster risk reduction and anticipatory actions. Created maps are then used to enable humanitarian actions like the Red Cross Red Crescent’s Forecast-based Financing (FbF) approach in areas most in need.

This study is conducted as Master Thesis for the Master of Disaster Management program at University at Copenhagen (supervised by Emmanuel Raju) and in collaboration with the Red Cross Red Crescent Climate Centre (supervised by Catalina Jaime). The research idea was proposed by the Red Cross Red Crescent Climate Centre as a contribution to ensure uncharted territories and the people who live there as well as first responders, humanitarian and development actors have access to complete, accurate and reliable geospatial data, to be used not only for emergency response but for anticipatory action and long term DRR and CCA planning. The development and set up of tasks on MapSwipe have been supported through the German Red Cross. The quality assessment of available OSM data in the study countries has been conducted and provided by the Geographical Institute of University Heidelberg.

This work is supported by the Danish Red Cross with funds from the Ministry of Foreign Affairs, Denmark.

Share this

PLEASE NOTE: CONTENT IS DISPLAYED AS LAST POSTED BY A PREVENTIONWEB COMMUNITY MEMBER OR EDITOR. THE VIEWS EXPRESSED THEREIN ARE NOT NECESSARILY THOSE OF UNDRR, PREVENTIONWEB, OR ITS SPONSORS.
SEE OUR TERMS OF USE