Better cooperation among military, civilian and humanitarian organizations can improve data management for aid and relief efforts
Joseph D. Martin
In a world plagued with communication challenges, the generation, validation and dissemination of information isn’t one of them. The use of digital sensors, commercial satellites and drones flood communications platforms with ones and zeros. Other technologies and human activities such as crowdsourcing may shape that information to address relevant questions and challenges, but often answers get lost in the sheer volume of data.
This is particularly relevant — and especially challenging — in humanitarian assistance and disaster relief (HADR). The nexus between civilian and military actors in this environment rates special attention due to altruistic and security-linked concerns. However, for any meaningful action to occur, military and civilian actors need to make significant changes in their trust and openness policies.
In a disaster, the militaries of the world are often the best-resourced organizations to respond. Be it communications, data collection, transportation or command and control, there is often no civilian analog with equally broad capacity. A downfall of this significant resource is that the military usually operates within the confines of its own protective barriers, be they physical or digital. Data are gathered, and information is shared in response to specific military needs with rare consideration for the utility of that same information outside the military.
Similarly, civilian or humanitarian organizations gather and share their own information to meet their specific needs during a response. Examples include other government agencies, private businesses, nongovernmental organizations (NGOs) and a range of international and regional organizations. Recent advances in technology even allow individuals and small-team, tech-driven players to use their unique skill sets to help others, often in an impromptu fashion.
Experiences relayed by practitioners during military events, combined United Nations and Association of Southeast Asian Nations gatherings, and even academic interactions point to a difference in collection means and capacities but a mutual willingness or ability to share. In the immediate aftermath of a large-scale disaster, information pours in from countless sources: militaries, local and foreign, gather imagery and intelligence data; local NGOs gather firsthand reports and on-the-ground images; and social media feeds populate Twitter, Facebook and other apps with real-time self-reporting. The volume of data continues to grow, and disaster-specific sharing mechanisms pop up based on the preferred local apps and data-sharing platforms.
Technologists from across the globe independently mine the metadata of the feeds and produce products accordingly. Intensity maps pop up, and various hot spots are triangulated and verified. These compiled products add to the growing stream of data, which, on the macro level, can be overwhelming.
However, at the micro level, challenges exist as well. In an instant, a U.S. surveillance aircraft on a HADR-sanctioned flight over the city captures an image of the destruction. The image is then scrubbed of any classified-related metadata and released to the affected country and is placed on a system such as the U.S. Department of Defense’s All Partners Access Network (APAN). Shortly thereafter, NGOs on the ground venture into that area to serve their primary benefactors.
Challenges to this scenario are extensive:
- With degraded communications, much of the technological analysis can be delayed in getting sufficient data and/or in relaying it to consumers in the field.
- Declassification and timelines for military assets may exceed the relevancy of the imagery taken.
- If the military imagery is scrubbed and becomes available, NGOs may not be willing to use it for fear of losing their humanitarian principles of neutrality, particularly in a combat zone. Or they may not trust the data if the source is masked and no longer verifiable.
With this flood of information — on a grand scale or with a single image — it is critical to realize that data and imagery are only useful if the consumer can do the requisite analysis. For example, with proper analytical support, an image of an affected urban area can serve the needs of search-and-rescue teams for damaged facilities and UNICEF and other children-focused NGOs on the condition of schools. The same schools can be used by shelter managers for displaced personnel. Engineering teams can determine the condition of roads, power lines, and water and sewer treatment plants. Medical personnel can determine how many beds are needed. Moreover, the details of population gathering points as well as the number of dead and injured can be ascertained. Six, 10 or 100 people can look at the same image and find answers to their respective questions if they can analyze the image for relevant information.
However, there are significant trust issues with data sharing and historical challenges such as those espoused in the military mantra of having a “need to know.” Many have expressed the advantages of transitioning to a “need-to-share” basis, but that is a separate discussion.
In the above urban image example, ground truth from resident NGOs and local providers can clarify many of the identified issues, but there are no known mechanisms to feed a central system. Additional types of data can augment the clarity by overlaying social media feeds, firsthand imagery and even crowdsourcing inputs. Unfortunately, even the idea of a central repository for disaster data is fraught with trust, technology and “not originated here” issues. A quick count shows 20-plus systems that are available in the open environment (of which APAN is one), with countless more at the individual level.
What is needed is a catalyst to drive the civil-military community, which is dedicated to saving lives and alleviating suffering, to expand its view on information sharing among a team of participants that range from single-person do-gooders to local and national first responders, NGOs and local and foreign militaries.
We are in a technological age in which that supposition is possible and sustainable. It is possible to create an information-sharing environment — beyond a mere platform — that enables trusted sharing not only of data but also of query capabilities to search for answers efficiently without duplicating imagery or data requests that others have already answered.
Inherent challenges start with trust and standards. Despite the humanitarian foundation of many civic-minded disaster relief groups, individual organizations are typically not inherently trusting of each other and certainly not of militaries or government agencies. Often, an NGO’s ability to generate sustainment revenue is based on its niche capability to serve a given need, and “giving away” that answer or data puts its existence at risk. Similarly, the government providers of data usually exist in a world of reduced trust to freely share data — especially where a perceived or real enemy can use that data to do harm.
How an entity would be able to share and integrate data, information and knowledge from a range of systems and then provide answers to an even larger number of queries presents a seemingly dauntless challenge. As big data and artificial intelligence expand, these challenges can be resolved. Perhaps a trusted organization, which is neither military nor humanitarian, can fill this gap by providing the expertise in information collection and analysis, while maintaining the trust of all involved.
The problem is not the lack of data. It is the lack of sharing, the lack of ability to analyze raw information, the lack of trust between the creators of the data and the organizations receiving it, and the lack of an organization to fill the gap among them. Each of these challenges has a solution, but it remains the responsibility of the community to decide when it wishes to tackle them.