Time-Varying Confounding Bias in Observational Geoscience with Application to Induced Seismicity

2025-10-18

AI Summary

This study addresses time-varying confounding bias in observational geoscientific data, which distorts causal inference—specifically in assessing the causal relationship between saline wastewater disposal and induced seismicity. We propose a longitudinal causal analysis framework grounded in the potential outcomes model, rigorously specifying identification conditions and systematically characterizing how time-varying confounders affect geophysical observations. Integrating parametric modeling with machine learning, we employ state-of-the-art doubly robust estimation techniques to correct for bias and obtain unbiased effect estimates. Validated on 2013–2016 observational data from the Fort Worth Basin, Texas, our framework quantifies, for the first time, the net causal effect of wastewater injection on seismic activity. It significantly enhances the reliability and interpretability of geological attribution from observational data, advancing causal inference in Earth sciences.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Security and Privacy: Data transparency and provenance

Abstract

Evidence derived primarily from physical models has identified saltwater disposal as the dominant causal factor that contributes to induced seismicity. To complement physical models, statistical/machine learning (ML) models are designed to measure associations from observational data, either with parametric regression models or more flexible ML models. However, it is often difficult to interpret the statistical significance of a parameter or the predicative power of a model as evidence of causation. We adapt a causal inference framework with the potential outcomes perspective to explicitly define what we meant by causal effect and declare necessary identification conditions to recover unbiased causal effect estimates. In particular, we illustrate the threat of time-varying confounding in observational longitudinal geoscience data through simulations and adapt established statistical methods for longitudinal analysis from the causal interference literature to estimate the effect of wastewater disposal on earthquakes in the Fort-Worth Basin of North Central Texas from 2013 to 2016.

Problem

• Addresses time-varying confounding bias in observational geoscience data
• Develops causal inference methods for wastewater-induced seismicity analysis
• Estimates causal effects of saltwater disposal on earthquake occurrences

Innovation

• Adapted causal inference framework for geoscience
• Used potential outcomes to define causal effects
• Applied longitudinal analysis for time-varying confounding