Leaf reflectance can surrogate foliar economics better than physiological traits across macrophyte species

Background Macrophytes are key players in aquatic ecosystems diversity, but knowledge on variability of their functional traits, among and within species, is still limited. Remote sensing is a high-throughput, feasible option for characterizing plant traits at different scales, provided that reliable spectroscopy models are calibrated with congruous empirical data, but existing applications are biased towards terrestrial plants. We sampled leaves from six floating and emergent macrophyte species common in temperate areas, covering different phenological stages, seasons, and environmental conditions, and measured leaf reflectance (400-2500 nm) and leaf traits (dealing with photophysiology, pigments, and structure). We explored optimal spectral band combinations and established non-parametric reflectance-based models for selected traits, eventually showing how airborne hyperspectral data could capture spatial-temporal macrophyte variability. Results Our key finding is that structural-leaf dry matter content, leaf mass per area-and biochemical-chlorophyll-a content and chlorophylls to carotenoids ratio-traits can be surrogated by leaf reflectance with normalized error under 17% across macrophyte species. On the other hand, the performance of reflectance-based models for photophysiological traits substantively varies, depending on macrophyte species and target parameters. Conclusions Our main results show the link between leaf reflectance and leaf economics (structure and biochemistry) for aquatic plants, thus envisioning a crucial role for remote sensing in enhancing the level of detail of macrophyte functional diversity analysis to intra-site and intra-species scales. At the same time, we highlighted some difficulties in establishing a general link between reflectance and photosynthetic performance under high environmental heterogeneity, potentially opening further investigation directions.

Automated summary

The study demonstrates that structural leaf traits like leaf dry matter content and biochemical traits such as chlorophyll-a content can be accurately represented by leaf reflectance, with errors of less than 17% across different macrophyte species. However, the performance of reflectance-based models for photophysiological traits varies significantly depending on macrophyte species and specific target parameters.