Deciphering climate-induced displacement in Somalia: A remote sensing perspective

This study is part of the Hummingbird project, funded by the European Commission in collaboration with VUB, and published in the PLOS ONE journal. 

Introduction

Since 1990, Somalia has experienced over 30 severe weather events, including 12 droughts and 19 floods, resulting in two major famines (in 1991–92 and 2011). These events displaced more than 3.2 million people. Approximately 70% of Somalis rely on agriculture and livestock, both highly vulnerable to climate change. The ongoing cycle of droughts and floods undermines food security and access to essential services, heightening vulnerability and driving displacement at various scales. Rapid climate changes significantly affect many aspects of life, often deteriorating living conditions in some areas to the point where residents are compelled to flee.

Despite increasing recognition of this issue, the dynamics between environmental factors and human mobility remain insufficiently explored. This study aims to investigate how advanced remote sensing analytics can be used to develop detailed climate indicators at a micro (district) level and to examine the relationship between climate factors and internally displaced persons (IDPs) in Somalia between 2016 and 2019.

Remote sensing technologies are particularly valuable for studying the relationship between climate change and human mobility. They provide critical environmental data at various scales, shedding light on the impacts of climatic factors, such as droughts and floods, on population movements. Satellite-derived data offer insights into climatic trends, including rainfall variability, temperature anomalies and land use changes, while also documenting environmental degradation.

Methodology

The GMV team used remote sensing datasets to create environmental indicators for Somalia, including a broad range of satellite data such as Sentinel-1, Sentinel-2, NASA DEM, Soil Moisture Index (SMI), Standardised Precipitation-Evapotranspiration Index (SPEI), and Google very high-resolution imagery, and non-satellite data like IDP records and official reports.

The GMV team developed a series of Python-based algorithms to process these datasets and generate various satellite-derived products, including: Agricultural Drought Indicator (ADI), Flood extent mapping, Land cover map and land cover change, Standardised Precipitation-Evapotranspiration Index (SPEI) anomalies, Soil Adjusted Vegetation Index (SAVI), Leaf Area Index (LAI), and Normalised Difference Vegetation Index (NDVI) and NDVI anomalies.

Findings

Doolow emerged as the most popular destination for displaced individuals, followed by Luuq. Conversely, Qansax Dheere had the highest number of departures due to drought events during the same period.

Our analysis identified 2017 as the driest year for Somalia, while 2018 and 2019 were marked by significant flooding. Regular rainfall periods, specifically the Gu and Dayer seasons, bring increased precipitation, which helps mitigate the impact of droughts during these growing seasons. In regions such as northern and central Somalia, where rapid-onset and slow-onset extreme climatic events occur either sequentially or concurrently, categorising environmental factors as either beneficial or detrimental is complex. For example, floods following severe droughts may help agricultural areas recover, but these environmental fluctuations do not always yield positive outcomes for livelihoods.

Flood events lead to immediate displacement, with little response time. People often begin evacuating just before the floods occur and frequently return home once the floodwaters recede, depending on the area’s recovery. Our study confirms that flood-induced displacement is largely localised, confined to specific districts, and typically remains within the same districts. In contrast, drought is a slower process, with displacement starting later and gradually increasing over time. Furthermore, returning after a drought takes much longer, as many displaced people likely perceive their original living conditions as unstable.

The effects of floods on displacement are more immediate than those of drought. Internal displacement follows soon after floods, and IDPs typically return home when floodwaters recede. Drought, however, is more gradual: people leave their homes with a delay, and their return is much more prolonged due to the severity of the conditions they face. A review of displacement patterns during the entire period shows that a district impacted by a severe drought does not always see a high level of displacement.

Despite challenges, our study offers an innovative approach to assessing environmental indicators and IDPs in Somalia. This methodology could be expanded to cover broader time frames and geographic areas. Future research might explore cross-border mobility and environmental factors. However, based on our findings and the challenges associated with available data, we assert that improving data quality and accessibility at both the national and international levels for regions most affected by climate change should be a priority. Finally, we suggest that triangulating quantitative and qualitative data will help illuminate the complex dynamics of climate-induced migration, laying the groundwork for future research in this field.

Author; Rahman Momeni