Unveiling environmental drivers of Moso bamboo sap flow using causal inference.

Abstract

Studying the relationship between Moso bamboo sap flow and environmental factors is essential for understanding the water transpiration patterns of this species. Traditional methods often rely on correlation analysis, but correlation does not imply causation. To elucidate the underlying mechanisms of how major environmental factors influence Moso bamboo sap flow, we analyzed the causality between them. First, the Fast Causal Inference algorithm was used to explore non-temporal causal relationships. Subsequently, the Latent Peter-Clark Momentary Conditional Independence algorithm was employed to further analyze the temporal causal effects. We found causal relationships among factors with low gray correlation coefficients. Besides, illumination, air, and soil temperature promote the density increase of sap flow, while carbon dioxide concentration, air humidity, and soil temperature inhibit bamboo sap flow density overall. Among these factors, illumination exhibits the longest lagged causal effect approximately around 80 minutes, whereas carbon dioxide concentration and soil humidity can quickly affect the sap flow density, with approximately 20 minutes. The study presents a novel methodological approach to analyze the complex interplay between environmental factors and sap flow, providing a more explanatory and logical framework. This study offers a novel methodological framework for disentangling the complex interactions between environmental variables and sap flow, providing deeper insights into the dynamic processes driving Moso bamboo water use. The findings contribute to advancing plant physiology and environmental science, while opening avenues for future research in related fields.