Causal Inference Framework for Ocean Microbial Community Responses to Warmer Temperature (Papers Track)
Minh Viet Tran (Helmholtz Munich, Ludwig-Maximilians-Universität München, Munich Center for Machine Learning); Christian L. Müller (Helmholtz Munich, Ludwig-Maximilians-Universität München, Munich Center for Machine Learning)
Abstract
Understanding how ocean temperature influences microbial communities is critical for forecasting ecological responses to climate change. This paper outlines a framework for applying the Rubin Causal Model (RCM) to observational amplicon sequencing data using matching methods. The aim is to estimate the causal effect of temperature on microbial taxa, while controlling for confounding environmental variables. This methodology offers a transparent, interpretable, and simulation-free way to extract causal signals from complex ecological datasets.