Climate Change AI Nordics

Climate Change AI Nordics

Nordic network for collaborations around tackling climate change using AI and machine learning.

Featured Paper: ONEKANA: Modelling Thermal Inequalities in African Cities

Authors: Sabine Vanhuysse, Angela Abascal, Stefanos Georganos, Jon Wang, Monika Kuffer

The new Climate Change AI Nordics Featured Paper is “ONEKANA: Modelling Thermal Inequalities in African Cities” by Sabine Vanhuysse and colleagues. This research addresses the pressing issue of thermal disparities in rapidly urbanizing African cities, where vulnerable populations are disproportionately affected by extreme heat due to environmental and socioeconomic factors.

The paper combines Earth Observation (EO) data, Machine Learning (ML), and Citizen Science to model near-surface air temperatures at a spatial resolution of 100m x 100m. By integrating satellite imagery with in situ temperature measurements, the paper provides detailed insights into how different urban areas experience varying thermal exposures. Target data comes from fixed sparse sensors, and crowd-sourced mobile measurements by citizens. The approach sheds light on urban heat distribution and may empower local communities to participate actively in data collection and climate adaptation strategies.

The findings underscore significant local differences in thermal exposure, emphasizing the necessity for targeted, cost-effective adaptation measures to protect the most affected urban populations. This work exemplifies the potential of combining AI-based tools with community engagement to address climate challenges in vulnerable regions.

For a comprehensive understanding, we encourage readers to explore the full paper (dipot.ulb.ac.be).

Published:

Categories:

You May Also Enjoy

November 2024

Published:

Welcome to the Climate Change AI Nordics Newsletter, November 2024! Read about recent and coming seminars, workshops, and publications from the network's researchers.

Frontiers in machine learning for weather forecasting

Event date:

Webinar with Joel Oskarsson, Linköping University. Recent years have seen rapid progress in using machine learning models for weather forecasting. These models show impressive performance, matching or even outperforming existing physics-based models, while running in a fraction of the time. This is fundamentally and rapidly changing the landscape of weather forecasting today. In this talk I will discuss the factors that enabled this paradigm shift, the core machine learning methods used and the research questions at the bleeding edge of machine learning for weather. In particular I will focus on how current methods can be extended to regional and probabilistic forecasting. For regional forecasting I will showcase graph-based methods for building limited area weather forecasting models. I will also discuss how generative machine learning methods can enable probabilistic forecasting, giving much-needed estimates of uncertainty and allowing for predicting extreme weather events.