Title: Explainable artificial intelligence-based approaches for climate change: a review

Authors: H. Can Barutcu; Serra Çelik; Murat Gezer

Addresses: Management Information System Department, Haliç University, Istanbul, Türkiye ' Department of Informatics, Istanbul University, Istanbul, Türkiye ' Department of Computer Science, Faculty of Science, Istanbul University, Istanbul, Türkiye

Abstract: Climate change is a significant problem that requires urgent action to identify and mitigate its causes. While artificial intelligence (AI) algorithms offer a promising tool to identify these causes, 'black box' constructs often obscure the meaning and impact of essential elements. At this point, explainable artificial intelligence (XAI), which illuminates algorithms and allows understanding of which factors significantly affect climate change, can be a saviour. This study focuses on applying XAI to reveal the factors affecting climate change, starting with identifying the areas that AI technologies can affect based on the existing literature. The pros and cons of artificial intelligence are discussed before delving into the concept of XAI and its potential in climate change research. This research aimed to clarify how AI can be effectively leveraged to address the complexities of climate change through XAI, highlighting the role of XAI in making AI insights into climate change understandable and actionable.

Keywords: explainable artificial intelligence; XAI; climate change; artificial intelligence.

DOI: 10.1504/IJGW.2025.145102

International Journal of Global Warming, 2025 Vol.35 No.2/3/4, pp.244 - 260

Received: 31 Oct 2023
Accepted: 17 May 2024
Published online: 19 Mar 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article