Climate

Tracing the space-time causal origins of Earth system extremes

Jhayron S. Pérez-Carrasquilla, J. Jake Nichol, Vanessa Robledo, Diana Bull, Katherine Dagon, Michael N. Evans, Maria J. Molina

Featured July 15, 2026

Original Paper

This analysis was generated by SciGrove. Upload your own PDFs or enter a DOI — and get the same AI breakdown on any paper.

Get started

AI-generated analysis — This is SciGrove's AI interpretation of the paper, not peer-reviewed content. Always refer to the original paper.

Simply

A new method called TraCE-ST acts like a detective, tracing backward in time through weather data to find out exactly which past events and locations caused a big storm or heatwave, even showing how much each factor contributed.

In depth

The paper introduces TraCE-ST (Tracer of Causal Evolutions in Space and Time), a novel framework that traces event-conditioned causal pathways in multivariate gridded Earth system data. It leverages existing local causal discovery methods but extends them into a Lagrangian-inspired approach to identify multi-step, multi-variable causal chains leading to extreme events, and probabilistically estimates the relative contributions of different drivers.

Key Takeaways

1
Introduces TraCE-ST, a probabilistic Lagrangian-inspired framework for tracing event-conditioned causal pathways in high-dimensional Earth system data.
2
Enables the identification of multi-step, multi-variable causal chains leading to extreme events, moving beyond local causal links to dynamic, event-centered explanations.
3
Quantifies the relative causal contributions of co-occurring and competing drivers through an ensemble-based probabilistic formulation, providing robust insights into complex phenomena.

Conceptual Flow

HIGH LEVEL

Methodology: Tracing Causal Paths

The method starts with a big event, then looks backward step-by-step to find what caused it, like following a trail.

Big Event

Trace Backwards

Past Cause 1

Past Cause 2

Past Cause 3

Results: Uncovering Multiple Drivers

It shows that big events often have many causes, and it can tell you which causes were most important.

Cause A

Cause B

Cause C

Combine & Influence

Big Event