Global Models and Local Knowledges

On machine learning and politics in flood forecasting


Author / Producer

Date

2023

Publication Type

Master Thesis

ETH Bibliography

yes

Citations

Altmetric

Data

Abstract

Climate change is expected to increase the frequency and intensity with which flooding occurs. Framed as an invading force (Brooks 2005), both public and private organizations seek ways to fight this harmful phenomena back. Regularly proposed in contemporary discourse as a potent weapon in this battle is artificial intelligence. The urgency and scale of the climate crisis calls for solutions with speed and reach. However, what is often obscured in grand visions of a sustainable future for all, is how we imagine and act to get there. With this project, I inquire into the politics of flood forecasting with machine learning. I study visions of the global with climate disasters, how these visions are operationalized with machine learning, and how such projects form together with identities, norms, and practices in specific localities. To understand how flood forecasting models are political, I ask what is being optimized, for whom, and who gets to decide.

Publication status

published

External links

Editor

Contributors

Book title

Journal / series

Volume

Pages / Article No.

Publisher

ETH Zurich

Event

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Science and Technology Studies (STS); Machine learning; Climate change; Flood forecasting; Situated knowledges; Politics

Organisational unit

09772 - Boenig-Liptsin, Margarita / Boenig-Liptsin, Margarita check_circle

Notes

Funding

Related publications and datasets