Data Driven Humanitarians are often short on time, resources and teammates. The requests for support they receive – or the needs they identify – are greater than the work they and their team (if they have one) can complete. Additionally, given the mixed understanding of the impacts of using Earth observation, they face challenges in advocating for more resources. In some larger NGOs, Data Driven Humanitarian are part of an organization-wide innovation or technology program, which can strengthen their work. In many others, they are lone actors, not situated within a broader country- or organization-wide structure.
While they have a strong understanding of the applications and limitations of Earth observation, many Data Driven Humanitarian have colleagues who don’t share this understanding. Their colleagues may not understand the diverse applications and impact Earth observation can have. Or, colleagues may appreciate the potential positive impact but have uneven understandings of the time required, technical complexity, and complex data landscape TSH’s must navigate to generate insights.
Accessing the high-resolution imagery required for humanitarian responses can be a fragmented, expensive and inefficient process—and a shared common understanding is rare. Further, integrating geospatial data into existing workflows and systems is challenging, particularly in terms of standardizing data collection, storage and analysis methods within and across organizations. Finally, access to EO data and the necessary technical support varies greatly across different national societies and humanitarian organizations. This disparity can hinder the effective use of EO data in less resourced settings.
Most of the Earth observation tools and datasets that Data Driven Humanitarian use were not designed for humanitarian contexts which are often marked by low internet/electricity connectivity, varying needs in terms of speed, rapidly changing contexts, geographical focus areas that are not always prioritized, and political/cultural sensitivities. Humanitarians also need rapid access to frequently updated imagery – something that is often unavailable or expensive. Each disaster, country and context is unique as well. Humanitarians must navigate assumptions that models trained on data from one region can be universally applied without contextual adaptation, leading to inaccuracies in different settings.
