ClimDetect: A Benchmark Dataset for Climate Change Detection and Attribution

2024-08-28

arXiv.org

AI Summary

This study addresses the challenge of improving detection and attribution (D&A) accuracy for anthropogenic climate change signals, enabling robust separation of human-induced forcing from natural climate variability. To this end, we introduce ClimDetect—the first standardized, open-source, and reproducible deep learning benchmark specifically designed for D&A tasks—comprising over 816,000 daily climate snapshots from multiple remote sensing and reanalysis datasets. Methodologically, we pioneer the application of Vision Transformers (ViTs) to climate fingerprint detection and propose novel spatiotemporal modeling and multivariate data fusion techniques tailored to climate dynamics. Experimental results demonstrate substantial gains in detection accuracy and cross-model comparability. All data, code, and trained models are publicly released, officially curated on Hugging Face, and designed to support community-driven evaluation and continuous benchmark evolution.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Neural Architectures and Foundation Models
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Detecting and attributing temperature increases due to climate change is crucial for understanding global warming and guiding adaptation strategies. The complexity of distinguishing human-induced climate signals from natural variability has challenged traditional detection and attribution (D&A) approaches, which seek to identify specific"fingerprints"in climate response variables. Deep learning offers potential for discerning these complex patterns in expansive spatial datasets. However, lack of standard protocols has hindered consistent comparisons across studies. We introduce ClimDetect, a standardized dataset of over 816k daily climate snapshots, designed to enhance model accuracy in identifying climate change signals. ClimDetect integrates various input and target variables used in past research, ensuring comparability and consistency. We also explore the application of vision transformers (ViT) to climate data, a novel and modernizing approach in this context. Our open-access data and code serve as a benchmark for advancing climate science through improved model evaluations. ClimDetect is publicly accessible via Huggingface dataet respository at: https://huggingface.co/datasets/ClimDetect/ClimDetect.

Problem

• Distinguishing human-induced climate signals from natural variability.
• Lack of standardized protocols for climate change detection studies.
• Enhancing model accuracy in detecting climate change signals.

Innovation

• Standardized dataset for climate change detection
• Integration of CMIP6 and reanalysis datasets
• Application of vision transformers to climate data