fbpx Correlation between climate anomalies and conflicts (UN Refugee Agency)

Climate anomalies and forced displacement

The problem

Millions of people are forced to leave their current area of residence or community due to resource scarcity and natural disasters such as droughts, floods, famine, and epidemics, which may influence the environment and exacerbate climate change (e.g. via deforestation to build new settlements, or crop destruction and fires). Our challenge partner, UNHCR, provides assistance and protection for those who are forcibly displaced, in this case internally displaced people (IDPs) inside Somalia.

The challenge goal

The challenge goal is to create a solution that quantifies the influence of climate anomalies on forced displacement and/or violent conflict through satellite imagery analysis. The challenge will involve the following tasks:

• Automate labeling of annotated satellite images of drought/flood conditions (e.g. NDVI)
• Use annotated satellite images to build a model(s) to predict displacement/conflict
• Use displacement/conflict to build a model(s) to predict climate/environmental changes

The datasets

• Satellite imagery: Landsat (EarthExplorer) and MODIS satellite imagery
Vegetation indexes (e.g.NDVI), Livestock, Hydrology, Settlement/shelters, Radar (e.g. detection of metal, trucks, tanks)
UNHCR-NRC forced displacement data
Conflict data: only violent conflict and fatalities
FSNAU Early Warning-Early Action Dashboard

Apply and be one of 35 collaborators and world-leading mentors to make a real-world impact.

Our Partners

UNHCR, the UN Refugee Agency, is a global organization represented in 134 countries dedicated to saving lives, protecting rights and building a better future for refugees, forcibly displaced communities, and stateless people.

Check our requirements below and apply at the end of the page.
  • Good English and access to a computer with internet
  • A good/very good grasp in computer science and/or mathematics
  • Student, (aspiring) data scientist, AI engineer, data engineer
  • Programming experience with C/C++, C#, Java, Python, Javascript or similar
  • Understanding of ML and Deep learning algorithms.
  • Availability of around 15-20 hours a week.
Your benefits:
  • Working with world-class AI mentors and domain experts
  • Being part of a driven and supportive community of collaborators in more than 60 countries
  • Certificates of participation and references
  • The chance to become a writer and be featured on our website and blog

Community Member (UNHCR)