Globally, the number of forcibly displaced people (FDP) reaches close to 90 Mio. A large share of FDP are internally displaced; not crossing international borders means no registration or protection status under UNHCR.
But then, where are those people in need? Earth observation (EO) products derived from satellite imagery play a key role in providing relevant and up-to-date information for humanitarian operations (Lang et al., 2019). Amongst the many benefits of remote sensing techniques in disaster- and conflict-related applications, timeliness and objectivity may be regarded as the most critical assets. This applies, for example, to dwelling extraction in refugee camp mapping or deprived urban areas for population estimation, where otherwise, no such figures exist at all, or convey – largely outdated – a distorted view. Recently, improved satellite sensor quality, data fusion techniques, and geospatial data availability in general have shifted the attention of researchers towards the automation information extraction process itself. In cooperation with Médicins Sans Frontières (MSF), one of the largest independent humanitarian organisations worldwide, our Christian Doppler laboratory GEOHUM develops AI-based algorithms and geospatial tools to support logistics, food, water and nutrition supply and public health interventions of MSF and other aid organisations. The talk delivers a critical view on the benefits (and challenges) we face with such technologies at stake, balancing lifesaving action and issues of sensitivity and privacy.
This online lecture is part of the EuMIGS Lecture Series 2022 on migration policy and refugee reception in the (trans-) local context.
Stefan Lang is Associated Professor of geoinformatics at Paris Lodron University of Salzburg.