Scholar iON
Academic Synthesis
This collection of scholarly works highlights critical themes in climate change research, focusing on mitigation, impacts, adaptation strategies, and public awareness. The IPCC report by Edenhofer et al. (2014) provides a comprehensive overview of strategies to mitigate climate change, emphasizing the role of policy and technological innovation. Burke and Stott (2017) analyze the impacts of anthropogenic climate change on the East Asian summer monsoon, identifying a decrease in monsoon rainfall and an increase in extreme rainfall events, which underscores the complex interplay between climate change and regional weather patterns. Costa et al. (2024) explore adaptation strategies using reinforcement learning to optimize responses to urban flooding, demonstrating the potential of advanced computational tools in climate adaptation. Zhou et al. (2019) address the challenge of public awareness by proposing evaluation metrics for generative models that visualize climate change impacts, bridging the gap between technological advancements and human perception. Collectively, these studies underscore the necessity of integrating scientific, technological, and societal approaches to effectively address climate change challenges.
The East Asian summer monsoon (EASM) is important for bringing rainfall to large areas of China. Historically, variations in the EASM have had major impacts including flooding and drought. We present an analysis of the impact of anthropogenic climate change on EASM rainfall in Eastern China using a newly updated attribution system. Our results suggest that anthropogenic climate change has led to an overall decrease in total monsoon rainfall over the past 65 years, and an increased number of dry days. However the model also predicts that anthropogenic forcings have caused the most extreme heavy rainfall events to become shorter in duration and more intense. With the potential for future changes in aerosol and greenhouse gas emissions, historical trends in monsoon rainfall may not be indicative of future changes, although extreme rainfall is projected to increase over East Asia with continued warming in the region.
Due to climate change the frequency and intensity of extreme rainfall events, which contribute to urban flooding, are expected to increase in many places. These floods can damage transport infrastructure and disrupt mobility, highlighting the need for cities to adapt to escalating risks. Reinforcement learning (RL) serves as a powerful tool for uncovering optimal adaptation strategies, determining how and where to deploy adaptation measures effectively, even under significant uncertainty. In this study, we leverage RL to identify the most effective timing and locations for implementing measures, aiming to reduce both direct and indirect impacts of flooding. Our framework integrates climate change projections of future rainfall events and floods, models city-wide motorized trips, and quantifies direct and indirect impacts on infrastructure and mobility. Preliminary results suggest that our RL-based approach can significantly enhance decision-making by prioritizing interventions in specific urban areas and identifying the optimal periods for their implementation. Our framework is publicly available: \url{https://github.com/MLSM-at-DTU/floods_transport_rl}.
With success on controlled tasks, generative models are being increasingly applied to humanitarian applications [1,2]. In this paper, we focus on the evaluation of a conditional generative model that illustrates the consequences of climate change-induced flooding to encourage public interest and awareness on the issue. Because metrics for comparing the realism of different modes in a conditional generative model do not exist, we propose several automated and human-based methods for evaluation. To do this, we adapt several existing metrics, and assess the automated metrics against gold standard human evaluation. We find that using Fréchet Inception Distance (FID) with embeddings from an intermediary Inception-V3 layer that precedes the auxiliary classifier produces results most correlated with human realism. While insufficient alone to establish a human-correlated automatic evaluation metric, we believe this work begins to bridge the gap between human and automated generative evaluation procedures.