Modeling Dwindling Water Resources from Imagery to Locate Instability in the Sahel

Global freshwater resources are dwindling due to many factors including population increase, agricultural practices, and climate change. As the water crisis worsens, the likelihood of global conflict could increase. The Sahel region of Africa is especially vulnerable to dwindling water resources, due to a harsh climate, poor infrastructure, and weak governance. To highlight areas of potential conflict, historical remote sensing data can be used to locate trends of water loss that can aid U.S. Government policy makers in allocating resources. Currently water resources are monitored using low-resolution imagery due to its historical libraries and availability. However, these low-resolution sensors do not provide the detail needed as water resources dwindle, becoming less detectable. To capture the true water resources available, deep learning methods on high-resolution imagery are needed. This study used high-resolution WorldView-3 imagery across the Sahel region to train deep learning models to use on images throughout the region. The model was able to classify pixels with an overall accuracy of 97% and an intersection over union of water compared to ground truth of 60%. Water loss estimates along with other geospatially enabled layers throughout the region were weighted using a geographic information system (GIS) multi-criteria decision analysis (MCDA) to highlight areas of potential instability throughout the Sahel.

Author

• Newton, Benjamin J

Language

• en

Keyword

• Water Extraction
• Regional Instability
• Remote Sensing
• GIS MCDA
• Sahel
• Deep Learning

Date created

• 2023

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Engineering Management

Degree

• D.Engr.

Advisor

• Ali, Daniyal A

Committee Member(s)

• Blackford, Joseph P
• Ali, Daniyal A
• Etemadi, Amir

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/3b591944n