Graduate Thesis Or Dissertation

Breaking Ground: Automating the Detection of Refugee Settlement Establishment and Growth through Landsat Time Series Analysis with a Case Study in Northern Uganda

Public Deposited

Downloadable Content

Download PDF


Attribute NameValues
  • Globally, there are 40 million internally displaced people (IDPs) and 25.4 million refugees as of 2018. Of this global refugee population, 19.9 million are under mandate by UNHCR, the United Nations Refugee Agency, while 5.4 million Palestinian refugees are protected by United Nations Relief and Works Agency for Palestine Refugees in the Near East (UNRWA). An additional 3.1 million people are asylum seekers. For the refugee population whose accommodation type is known, over a third are living in settlements managed and planned by UNHCR. Since 2015, Uganda has welcomed a massive influx of over 700,000 refugees from South Sudan, Democratic Republic of the Congo, Burundi, and other East African nations, and currently hosts over 1.2 million refugees with 92% of that population living in UNHCR managed settlements. Progressive refugee policies in Uganda have led to the establishment of a network of rural UNHCR-managed settlements to host the rapid influx of refugees. Despite these settlements being essential spaces for physical protection and humanitarian aid distribution and reception, the sheer rate of population influx and settlement growth introduces uncertainties around sustainable site planning, aid delivery, food security constraints, and environmental change. For the vast majority of settlements, refugee communities, aid agencies, and host country governments do not have accurate information on settlement establishment, land use conversions, settlement growth, structural densification, and broad-scale changes in spatial arrangement and land cover over time. Collectively, the absence of this information yields a knowledge gap around refugee settlement typologies that would benefit settlement-level decision making and also support systematic comparison between physical settlements and refugee settlement processes. The goal of this thesis is to characterize the spatial and temporal patterns of refugee settlement establishment and growth by using Landsat satellite imagery and an automated disturbance detection approach with the case study of the Pagirinya Refugee Settlement in Northern Uganda, settled in mid-2016. The findings of this thesis suggest that the Pagirinya Refugee Settlement was established very quickly (within one month) with first road delineation followed by rapid onset of built-up area, which is then followed by the conversion of natural grassland to small-scale agriculture approximately six months later. The method yields an exceptionally high accuracy of 87.5% for detecting land cover disturbances associated with settlement land use, with a median 3.5 month temporal lag in identifying the timing of establishment and growth. This proof of concept paves the way for developing a near real-time establishment and growth monitoring tool, which can aid in refugee response and evaluation efforts.
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Committee Member
Academic Affiliation
Rights Statement
Peer Reviewed



This work has no parents.

In Collection: