With funding from DG ECHO, FSD is spearheading an initiative on Drones in Humanitarian Action in collaboration with CartONG, Zoï Environment Network and UAViators. The objectives of this meeting were to provide a progress update on the use of drones in emergency response, to discuss new case studies and to consolidate findings and outputs. The overall aim of this initiative is to provide humanitarian organisations with a roadmap for the effective and safe integration of this technology in support of more efficient and effective delivery of humanitarian assistance.
A 2-day consultation meeting was convened on the use of drones for mapping in humanitarian contexts. The purposes were to:
• Offer a platform for constructive discussion to take stock of how drones and related technology are useful in creating maps in humanitarian contexts
• Identify how best to adopt and roll out this technology for humanitarian organisations
• Discuss the way forward, including next steps towards safe, effective and efficient integration of drones for mapping in humanitarian work
The 40 participants included technology experts and representatives of humanitarian NGOs, governments, donors and two drone manufacturers. The results of this discussion will provide the basis for a Guideline Publication on the use of mapping drones in humanitarian action, which will be published later in 2016. Case studies covering past drone deployments in humanitarian contexts were provided as background reading for the meeting (see Annex I).
The consultation was conducted over a day and a half and comprised sessions on the topics of:
• Case Studies and Existing Practice
• Operationalisation by Humanitarian Organizations
• Mapping Drones and Alternatives
• Data Sharing and Distribution
• Emergency Response and Coordination
Other presenters on cross-cutting themes shared their viewpoints during open discussion sessions.
Case Study Presentations
The case studies describe the use of drones for mapping in camp management and in support of recovery after disasters, especially in urban areas. They are also examples of drones being used in support of disaster preparedness. Some of the case studies covered pilot projects, whereas other case studies described fully operational uses of drones. Highlights and key findings include the following: Local capacity-building is a priority in many drone projects. Project planning and roll-out consider local capacity, and projects often include a university or a mapping and open technologies community such as OpenStreetMap. Communicating with communities is important when using drones. Deployments were rarely met with local community opposition, possibly because project staff briefed communities on the use of drones and the objectives of the projects, and often included them in the mapping process. Communities easily accepted the small drones, and are likely to prefer them to larger drones. In post-conflict settings, terminology is important. Especially in post-conflict settings, the use of alternative terms for “drones” such as “aerial imaging device” or “UAV” was helpful in engaging authorities for the necessary permits. Showing the drone physically (especially if it is a lightweight model) to authorities can help facilitate collaboration. Different resolutions were used depending on the type of area mapped.
Many of the uses involve dense urban environments that called for higher resolution than mixed urban/rural environments. In urban areas, using drones for mapping had benefits that extended beyond providing a high quality map. The high quality imagery collected by the drones enabled the project staff to detect changes in an urban area following a disaster. This additional facility could be used to evaluate damage and improve humanitarian response activities. Project team sizes vary greatly. For larger projects, including land tenure and census projects, a whole team of GIS and flight specialists had to be built around collecting, managing and evaluating drone imagery data. The size of the required staffing support is a consideration when deploying drones in support of disaster response.
Drone technology has evolved sufficiently that automated flight paths and safety features facilitate the basic piloting of drones with little training. The focus and challenge is shifting to processing, analysing, storing and using the data collected during flights. Data processing is often several times more time consuming than the flights, and involves management of large data sets. Several of the case studies identify the data management component as one of the main areas of focus in the deployment of drones for humanitarian mapping.
Choosing the right context for a drone mapping project is important. Most humanitarian NGOs mentioned that they are only exploring their use in non-conflict settings (for now). Organisations are reluctant to deploy due to reputational risks, such as possibly jeopardising their neutrality status. Some also expressed concern over the risk of losing operating licenses in particularly sensitive contexts.
Other concerns and suggestions on the operationalisation of drones included the following points:
The right tool: It is important that partners verify that drones are the best tools to use before starting a project.
Communication and transparency: Drones collect information in an asymmetrical fashion: the operator sees more than the observed population sees. To ensure transparency, it is important whenever possible to share maps with the local authorities and communities being observed and mapped.
Safety: Lightweight microdrones are the most commonly used models in humanitarian settings and are unlikely to cause third party damage or injury.
Perception: Very small, stealthy drones may not be appropriate for use in humanitarian activities at this point, as they could be perceived as gathering intelligence.
Expectations: Expectations on what can be achieved through the use of drones needs to be managed.
The demand side: UAVs in the humanitarian sector have gone through several years of a sort of experimental mode characterised by supply-driven activities aimed at generating interest in drone applications and at raising awareness of operational scope. As UAVs prove their value and as regulations on their use start to become clearer (though there is still much work to do), the humanitarian community should start to compile case studies in order to better understand what to specifically demand of UAVs in what context.
Coordination: Some actors think that duplication of effort in assessments and limited information sharing are common in the humanitarian community. Operationalising UAVs should take into consideration the need for establishing a drone coordination task force that would potentially centralise drone deployments, reduce multi-organisation efforts, share drone imagery and expertise with humanitarian organisations responding to a crisis, and consolidate accountability to affected populations through the use of this technology.
Emergency Response Drone (deployment within 72 hours of a disaster)
Not much documentation (including case studies) is available on the use of drones in mapping in the immediate response phase to a disaster. Indeed, as of 2010, the average number of days between a disaster and any type of robot deployment was 6.5 days. The only known case of use of drones in the immediate response phase was after Hurricane Sandy in Haiti in 2013. At the time, IOM had an existing drone capacity and also a trained emergency response team in-house, so deployment was simplified. While there is a lot of theoretical information on how drones could contribute, the real impact or utility of mapping drones in this capacity has not yet been demonstrated in the field.
Current data processing, which is paramount to the timely and effective delivery of aid in the first stages of emergency response, is not fast enough to add value to tactical decision-making. Drones must enable emergency responders in the field to make better decisions faster to deliver added value.
Standard operating procedures
No global mandate for coordination of drones is in place yet but the World Food Programme may be in a position to develop a framework that would allow for global and local stand-by capacity. This framework would need to include the coordination among actors on the ground and with host government authorities for permits, preparedness, prepositioned equipment and tools to support the work associated with UAV deployments.
Regulations are a major impediment, both to local international capacity in deploying within the first 72 hours of an emergency, and there is a need for streamlined procedures to allow for rapid response.
To ensure that drones can be deployed following a natural disaster or in other emergencies, local actors trained on the humanitarian response workflow must be available through external rosters of professionals.
Drones vs. Alternatives
Satellites, airplanes and helicopters are the three most commonly used alternatives to drones in aerial mapping and assessments. In some instances the available outputs may overlap, but for the most part, these tools should be seen as complementary to one another. Satellite imagery resolution typically ranges from 30 cm to 1.5 m and has minimal coverage of 25 km², whereas drone imagery is mainly used around 3.5-15.0 cm resolution, and most fixed wing drones in use up to now for humanitarian purposes can cover up to 2 km² in one flight of 40 minutes. The images in Figure 1 demonstrate how drone imagery provides a clearer picture compared to that of a satellite.
Drones are able to capture images beneath a cloud cover, and thus provide additional flexibility compared to satellites. At present levels of technology, mapping drones are most apt to provide the following:
• Ultra-high-resolution imagery as commonly required for cadasters, urban planning and certain camp management requirements
• Time-sensitive imagery in regions with frequent cloud cover
• Imagery for the digital modeling of 3D surface elevations Drones are most useful in locations where access is not an issue as flight times and distances are limited.
To facilitate information sharing concerning regulations in specific countries, the existing UAViators regulations database has been enhanced by FSD via the ECHO-funded project. All users are welcome to contribute and to suggest how it could be made more user-friendly.
Data Sharing and Distribution
Managing, sharing and distributing data are now the main challenges when working with mapping drones. Several systems for sharing and managing data came up during the discussion, including OpenAerialMap, MapBox and GeoNode. Google Earth can be used to share images with a non-technical audience by importing the kmz files. Dropbox has been useful for storing and sharing files. In many advanced uses, specific GIS skills are needed to evaluate and process the mapping imagery. A presentation was also made of SnapBird, a new service to support tasking and coordination of aerial footage.
Participants of this meeting agreed that the humanitarian community is at the advent of the use of drones in humanitarian responses. Many believe the use of drones will increase in the coming years as the technology is finally ready to adapt to the needs of humanitarians and to provide the added value required to facilitate the delivery of aid and the provision of assistance, and to ensure stronger data-driven analyses of humanitarian needs.
Drones have been tested and proven successful in two ways: 1) for long-term implementation, and 2) when deployed a few weeks following a disaster, during the recovery phase. A third use – the use of drones in the direct aftermath of a disaster – is emerging as well, but as discussed, there are several additional components that need to be put in place before this can be institutionalised.
A stronger knowledge base on how and when drones can be effective for mapping compared to traditional methods is required. An understanding of how local authorities and communities perceive the use of drones in the context of humanitarian response is important, as is the ability to manage perceptions.
On the technical level, the main issues are the management and sharing of data, including the time it takes to analyse the data. The need for developing local capacity and potentially further building a network of external experts (a roster) has been stated by many as a possible next step.
Prepared by: Denise Soesilo, Audrey Lessard-Fontaine, Jessica DuPlessis
Annex I: List of Case Studies
The timely availability of aerial imagery of the damage caused by Hurricane Sandy, combined with existing open source imagery and census data, allowed analysts to determine with precision where assistance was needed. The ability of drones to acquire the imagery depended on pilots gaining access to the affected areas, but did not require the clear skies necessary for the effective use of satellite imagery, and the drones accomplished the task seven days before the satellites.
A project employing drones in the wake of Typhoon Haiyan explored how aerial imagery might support recovery and reconstruction activities. Ultimately, the imagery captured by drones became useful in both a tactical and strategic sense during the retrofitting of shelters, and helped not only to identify and verify the shelter sites, but also to help determine the placement of latrines. The mission provided a rich learning experience on the operational use of aerial robotics in a disaster recovery context.
Heavy floods and landslides in Bosnia Herzegovina shifted minefields and explosive remnants of war (ERW) into inhabited areas. The Belgian Royal Military Academy (RMA) team worked with the Bosnia and Herzegovina Mine Action Centre (BHMAC) to use drone images that would help model the potential locations of some of the many displaced ERWs and mines. These models then were used to narrow down the search radius for demining teams.
Drones provided high-resolution, up-to-date imagery that was essential to the development of exposure maps and for modelling flood risks in Dar es Salaam. The success of the project has led to further applications and to government funding for infrastructure improvements in vulnerable areas identified by the project.
The aims of the project were to provide drone training to locals around Kathmandu and to demonstrate the potential use of drone imagery in disaster response. To this end, relatively affordable consumer drones, together with mapping software and a flight-planning app, were used to achieve similar results to those a professional mapping drone would produce.
IOM considers the use of mapping drones a helpful tool to plan shelter units and to moni- tor the evolution of camps. As a result, IOM has been using drones since 2012 for this pur- pose. An initial drone flight in Port-au-Prince in February 2012 provided figures on seven camps for persons displaced by the 2010 earthquake. The imagery was used to delimit the extent of the camps, and then to count and uniquely identify the tents within each camp. These data could then be linked to IDP data stored in the Displacement Tracking Matrix (DTM). This helped establish the size of camps (as the number of tents and number of people were known), organize the camps into blocks and zones to inform registrations and operations for decision making.
The International Organization for Migration, in collaboration with the National Statistics Office of Haiti, conducted a census of areas and populations affected by the 2010 earth- quake. Insufficiently precise GPS and out-dated reference imagery were inadequate to the task of clarifying land tenure status. IOM therefore conducted drone flights to obtain the high-resolution imagery essential to the preparation of assessments in Haiti’s densely populated slums. The drone imagery enabled the preparation of precise maps of the enu- meration areas, and the maps enabled the enumerators to pinpoint exactly which build- ings to assess and to link individual buildings to their owners. Local support included vol- unteers from the Haiti OpenStreetMap community, who contributed directly to the project by digitizing buildings and roads.
Annex II: List of Participants
CartONG – Audrey Lessard-Fontaine
CartONG – Martin Noblecourt
Consultant/MSF – Louis Potter
Delair-Tech – Julie Populus
ECHO – Daniel Clauss
ESPI – Elisabeth Veit
FSD – Hansjoerg Eberle
FSD – Valeria Fabbroni
FSD – Denise Soesilo
FSD/Aurea Imaging – Bert Rijk
GICHD – Inna Cruz
Handicap International – Pierre Gallien
Handicap International – Paul Vermeulen
Human Rights Watch – Josh Lyons
ICRC – Sebastian Langer
IIHA Fordham University – Giulio Coppi
IMMAP – Chris Habib
IMPACT – Renaud Zambeaux
IMPACT – Vincent Annoni
IOM – Sebastian Ancavil
MapAction – Colin Spiller
Medair – Joel Kaiser
MSF OCG – Frederic Ham
MSF OCB – François-Xavier Dosogne
MSF OCG – Philip Janssens
OpenAerialMap/Development Seed – Nate Smith
Potentiel 3.0 – Frederic Moine
SenseFly – David Rovira
SnapBird – Olivier Cottray
UN OCHA – Manuel Timmermans
UN OCHA/FIS – Yaëlle Link
UNHCR – Julia Chatellier
UNITAR-UNOSAT – Manuel Fiol
UPS Foundation/GAVI – Kevin Etter
WeRobotics/UAViators – Patrick Meier
World Bank Group – Edward Anderson
World Food Programme – Haidar Baqir
Zoï Environment Network – Christina Stuhlberger
Zoï Environment Network – Otto Simonett
Drones in Humanitarian Action is partially funded by DG ECHO. This website covers humanitarian aid activities implemented with the financial assistance of the European Union. The views expressed herein should not be taken, in any way, to reflect the official opinion of the European Union, and the European Commission is not responsible for any use that may be made of the information it contains.