Causality for Earth Science -- A Review on Time-series and Spatiotemporal Causality Methods

TL;DR

This survey provides a comprehensive map of time-series and spatiotemporal causality methods in Earth science, detailing causal discovery and causal inference, their underlying assumptions, evaluation metrics, datasets, and tools. It catalogues a broad spectrum of methods—ranging from Granger causality and PCMCI to TiMINo, TCDF, and STCINet—alongside spatiotemporal extensions like Mapped-PCMCI, ICSD, and STCD, and discusses their Earth-science applications. The paper also highlights key challenges (e.g., nonstationarity, latent confounding, data gaps) and open questions, offering guidance on when to apply specific methods and how to validate results using synthetic and real-world benchmarks. By compiling datasets and software resources, the authors provide a practical primer for researchers aiming to apply data-driven causality to climate, hydrosphere, cryosphere, and related Earth-system questions, while outlining opportunities for methodological innovation and interdisciplinary collaboration.

Abstract

This survey paper covers the breadth and depth of time-series and spatiotemporal causality methods, and their applications in Earth Science. More specifically, the paper presents an overview of causal discovery and causal inference, explains the underlying causal assumptions, and enlists evaluation techniques and key terminologies of the domain area. The paper elicits the various state-of-the-art methods introduced for time-series and spatiotemporal causal analysis along with their strengths and limitations. The paper further describes the existing applications of several methods for answering specific Earth Science questions such as extreme weather events, sea level rise, teleconnections etc. This survey paper can serve as a primer for Data Science researchers interested in data-driven causal study as we share a list of resources, such as Earth Science datasets (synthetic, simulated and observational data) and open source tools for causal analysis. It will equally benefit the Earth Science community interested in taking an AI-driven approach to study the causality of different dynamic and thermodynamic processes as we present the open challenges and opportunities in performing causality-based Earth Science study.

Figures (7)

• Figure 1: Three structures of causal graphical model. The nodes represents entities and edges represent direction of causal relation.
• Figure 2: An example to demonstrate causal sufficiency.
• Figure 3: An example to demonstrate the causal Markov condition/property.
• Figure 4: An example to demonstrate the acyclicity property where the graph in (a) holds acyclicity and (b) violates acyclicity.
• Figure 5: An application of spatiotemporal causal discovery method (Mapped-PCMCI) for identifying teleconnections. tibau2022spatiotemporal