Abstract: A crucial aspect of plant ecology is identifying the mechanisms that drive the diversity of strategies for growth, reproduction, and survival, in particular in the Amazon Forest, which has the highest tree diversity in the planet. Among plant traits, Leaf Mass per Area (LMA) and Wood Density (WD) are classified as "soft traits" because they are morpho-anatomical traits that are expected to be associated with function. In the economic spectrum framework, species that are acquisitive have lower values of LMA and WD, while those that are conservative have higher values. Higher LMA is linked to higher photosynthetic rate on a global scale. In contrast, hydraulic traits are considered "hard traits" as they are physiological, and enable predictions about tree mortality and the associated impacts on forest carbon and water cycling under climate change. Unfortunately, hydraulic traits are difficult and time consuming to measure. High throughput indicators of hydraulic traits would allow rapid assessments of forest-scale drought vulnerability. While the relationships between leaf reflectance spectra and LMA have been well-established across various biomes, the same cannot be said for wood density and hydraulic traits, particularly in the Amazon. This raises the question: can leaf spectra effectively predict hydraulic traits in the Amazon?
We sampled 254 individual Amazonian trees for water potential, LMA and measured leaf reflectance for the same leaves, and performed hydraulic percent loss conductivity curves (i.e., P50) and measured wood density to the same branches. We examined relationships between leaf spectra and hydraulic traits through remote sensing indices and partial least square regression (PLSR).
The study found that leaf spectra had a strong power of predicting LMA (R2 > 0.8 , RMSE < 10 g/m2 ), leaf water potential (R2 > 0.5 , RMSE < 1.48 MPa ), and wood density (R2 > 0.5 , RMSE < 0.11 g/m3 ), but not a strong power for P50, based on PLSR. The water absorption bands of the leaf spectra showed high variable influence on projection scores, and the Water Band remote sensing index also predicted leaf water potential. However, wavelengths in the visible domain were also influential and warrant further investigation.
