Predictability of leaf morphological traits for paleoecological reconstruction: the case of leaf cuticle and leaf dry mass per area

Our power to predict the future relies on our knowledge of the past. Paleoproxies are a powerful tool for understanding environmental and ecological conditions and changes across different time periods. However, constructing a functioning paleoproxy requires a well-constrained and robustly tested model. This is challenging, especially if ecological traits are involved. In the current study, we constructed an extended data set to test the reliability of the derivation of leaf dry mass per unit area (LMA) from the thickness of fossil gymnosperm cuticle. Specifically, we tested whether different leaf types (broad leaves, needles, and scales), intraspecific variability in cuticle thickness, and growing conditions affect the functioning of the proxy. Taxonomic groups were analyzed to uncover the possible taxonomic influence on LMA, cuticle thickness (CT), and the LMACT relationship. Our results indicate that the CT versus LMA relationship depends on multiple factors that can have various and incongruous effects on this relationship, depending especially on leaf type and growing conditions. We conclude that CT measured from gymnosperm fossils could be used as a proxy for LMA in past ecosystems for some broad- and scale-leaved gymnosperms but not for needle-leaved gymnosperms. However, caution must be taken when comparing species from different environments or growth conditions.