Development and aid professionals are increasingly exploring how to use satellite data to achieve goals in a variety of sectors. While data could help humanitarians coordinate the best response to a natural disaster, public health organizations use it to track and control the outbreaks of diseases such as malaria. Urban planners may use maps to help plan and plot refugee camps, and NGOs to build wells and sanitation facilities in close proximity to the settlements that need them most.
While it’s important to explore the various applications of satellite data, it is also vital that stakeholders consider how to collect and use that data ethically, according to Rupert Allan, former country manager of Humanitarian OpenStreetMap Team, or HOT, in Uganda.
“Mapping is an inherently colonial activity ... using a god-like satellite to take images from above, and using that to decide how resources should be distributed.”— Doug Specht, senior lecturer at the School of Media and Communications, the University of Westminster
In particular, according to Allan, it’s important to consider the power dynamics inherent to the collection of satellite data in emergency and development contexts, including those related to access, control, and ownership.
“I’m not sure you can talk about the use of satellite imagery and digital technology in developing countries without talking about the premise of old colonial patterns, and the problematics of white males telling subordinated ‘beneficiaries’ how to do things better,” he said.
To avoid that, Allan recommends that organizations think about who is directing and controlling the collection of data, and to what degree it’s actually “accessible and accountable to and by local communities.”
Access, control, and ownership
While satellite data has many potential applications in development, if used carelessly there’s a real danger of alienating local partners, according to Doug Specht, senior lecturer at the School of Media and Communications at the University of Westminster.
In-depth for Pro subscribers: Tips for becoming a data-driven organization
At Devex World, experts discussed different approaches to becoming a data-driven development organization, highlighting good data practice as well as barriers in the development sector.
“In the development context, there’s been a lot of talk about the importance of being participatory, about whose voice is being expressed,” he said. “But mapping is an inherently colonial activity, and there’s nothing less participatory than using a god-like satellite to take images from above, and using that to decide how resources should be distributed.”
Another potentially thorny ethical issue relates to the financial incentives created by satellite technology: because satellites are so expensive to operate, most satellite companies are commercial operations, with commercial interests. While development professionals can get access to data through public-private partnerships and CSR initiatives, those financial incentives still affect what data is available, he said.
“At the end of the day you are never just using the data, you are also beholden to whoever owns the satellite,” he said. “It’s not that they explicitly tell you how to use the data — it’s that they only collect certain types of data due to their own needs, and that then impacts the types of projects that NGOs design.”
Consider keeping data open
To avoid these dynamics, Allan says it’s vital that stakeholders think carefully about who controls the collection of data, what platforms are being used to store and organize the data collected — and if those platforms are proprietary, or open-sourced — and what will happen to the data after the project is complete.
“The problem with proprietary data is that each NGO might be mapping different things, and they aren’t necessarily sharing that data,” Allan said. “So you might have Oxfam mapping one type of water source, but ignoring WaterAid, who may be mapping another, potentially leading to false resource reports.”
Another issue with proprietary data is that after the project is complete, it often becomes inaccessible to anyone except the NGO and donors, he said.
“The data is not only getting wasted, but it’s also potentially insecure, through lack of sustained ownership,” he said. “One way to get around this issue is to make data open-source, available in the public space, accountable to — and also update-able by — local communities.”
Not only does this ensure that data is accessible and useful to local partners both during and after the project is complete, but it can also address the issue of “dark data”— that is, data that international NGOs might not consider important or know they ought to collect.
“Dark data is missing data due to epistemic and ideological assumptions of powerful constituencies,” Specht explained. “You’ll find that communities whose knowledge systems don’t fit into Western discourse are deemed too difficult to collect or map.”
As an example of how to combine open-sourced mapping technologies with local knowledge and consent, Allan points to a project by HOT in Uganda to map out water points in refugee settlement areas that uses this OpenStreetMap platform, allowing community members to create “home-made geospatial data,” layering data surveys and points over community-traced satellite imagery of infrastructure photogrammetry.
To ensure comprehensive and complete mapping, community members are trained and mentored to map out all water access points — both functional, and non-functional — and plot them on open-sourced HOT maps on their smartphones. By reporting each and every source of water, formal and informal, they were able to see where households actually gathered their daily water — which wasn’t always an “official” source.
This hyper-local context helped development professionals better understand how to best allocate important resources for humanitarian interventions. Mapping nearby toilets and sanitation facilities also allowed teams to cross-reference water sources that might become contaminated during floods, and to identify nonfunctional wells that needed to be repaired. For Allen, the fact that the mapping technology was open-sourced — and thus accessible to community members — was critical to the program’s success.
According to Allan, HOT Uganda is now the biggest contributor to the United Nations Office for the Coordination of Humanitarian Affairs Humanitarian Data Exchange for Uganda, and has become the go-to resource for geospatial analysis and intervention in the country.
“OpenStreetMap is more up-to-date and accurate than Google maps, because it is potentially more interactive, publicly-owned, and doesn’t favor the commercial — it’s about community concerns,” he said. “This allows for a more conversational and dynamic approach to mapping accessible to local stakeholders, rather than a snapshot that’s fixed in time.”
Consider the unintended consequences
Unfortunately, there aren’t easy answers to the ethical questions that are evoked by the use of satellite data.
“One of the key issues with satellite technology is that it upsets the power dynamic,” Specht said. “There can’t be any parity of power when one party has access to technology that tells them everything about your daily movements, and the other side has nothing. So you’re not starting on a level playing field.”
Even so, Specht recommends that stakeholders planning to draw on the power of satellite technology think critically about issues like unequal power dynamics, ownership, and potential unintended consequences, and how they can engage local actors in a more collaborative and empowering way.
“Used unilaterally, satellite technology can erase local voices, by luring us into thinking we have enough data to answer critical questions,” he said. “To make this process participatory, we need to find a way to bring in those local voices, and to recognize that power dynamic.”
Visit the Data for Development series for more coverage on practical ways that satellite data can be harnessed to support the work of development professionals and aid workers. You can join the conversation using the hashtag #DataForDev.