Q&A: Satellite imagery reveals what casualty figures alone do not tell us
How do we form a picture of conflicts? A new study published in the journal Nature explores how different types of data can be combined to give a fuller understanding of armed conflicts. Lead author Valerie Sticher uses the examples of Ukraine and Myanmar to explain the advantages and limitations of automatically analyzed satellite data.
This study in the journal Nature examines how satellite imagery and text-based data can provide a more comprehensive understanding of armed conflicts beyond just casualty figures. The research highlights the advantages and limitations of combining different data types, using examples from Ukraine and Myanmar. While casualty figures are crucial, they do not capture other consequences like displacement or destroyed livelihoods.
Satellite data from conflict regions has been used before, but automatic analysis allows for systematic data collection. The study combines text-based and satellite-derived data from Ukraine and Myanmar to reveal how conflict dynamics evolve spatially and over time. In Myanmar, text-based sources suggest massacres against the Rohingya occurred in the first week, while satellite data shows violence continued for months.
In Ukraine, damage from Russian territorial gains was more extensive than during Ukrainian recaptures. Satellite data can only capture events with visible physical effects, so it cannot identify perpetrators, displacement patterns, or sexual violence. Combining different types of data is essential for a more complete picture of conflicts and for targeting humanitarian aid effectively.
However, the models used for analyzing satellite data need diverse training data to work reliably in various conflict regions, particularly regarding conflict-related sexual violence, which is inadequately represented in text-based data.
Written by urgent.news from Phys.org's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
