Advances in causal discovery methods for ecological time series.

Area of Science:

• Ecology
• Complex Systems Analysis
• Data Science

Background:

• Advances in data collection yield extensive ecological time series.
• Dynamic ecosystems require sophisticated causal inference methods.
• Granger causality (GC) and convergent cross mapping (CCM) are key dynamical causal discovery techniques.

Purpose of the Study:

• To review and synthesize foundational concepts and recent developments in GC and CCM.
• To explore the strengths, limitations, and interrelationships of these causal discovery methods.
• To highlight the applicability of temporal causal discovery methods to ecological data.

Main Methods:

• Review of Granger causality (GC), transfer entropy (TE), and convergent cross mapping (CCM).
• Synthesis of recent advancements in temporal causal discovery.
• Analysis of method applicability to ecological time series data.

Main Results:

• GC and CCM offer valuable insights into nonlinear and chaotic systems.
• Recent temporal causal discovery methods, though less familiar to ecologists, show promise.
• A clear understanding of these methods' interrelationships and limitations is crucial.

Conclusions:

• A rigorous framework for time-series-based causal inference is needed in ecology.
• Increased awareness and application of advanced causal discovery methods can foster deeper ecological understanding.
• This review encourages further research and development in ecological causal inference.