Linking Riparian Vegetation Diversity and Seasonal Spectral Greenness in Temperate Atlantic Rivers

ABSTRACT Aims Riparian ecosystems host a wide array of plant species and functions, resulting in different primary production patterns that may be inferred with remote sensing indicators. We aimed to assess the relationship between monthly spectral greenness and vascular plant biodiversity in riparian forests using Sentinel‐2 images. Location Cantabrian region, Spain. Methods We evaluated the effects of six indicators of riparian vegetation composition and functional diversity (Gini–Simpson; species richness; plant height range; specific leaf area range; and the relative abundance of evergreen and exotic species) on two Sentinel‐2‐derived spectral greenness indices (Enhanced Vegetation Index, EVI; Normalized Difference Vegetation Index, NDVI) using linear mixed models, adding the elevation and the month as additional predictors, and the year as a random factor. Results We found that all of the indicators had significant positive effects on spectral greenness, with Gini–Simpson being the strongest predictor, followed by plant height and specific leaf area ranges. These effects varied throughout the year, with stronger influences on NDVI during the dormancy season and on EVI during the growth and peak‐production season. The abundances of evergreen and exotic species were weaker predictors, yet significant. Conclusions The results confirm the significant relationship between riparian vegetation greenness and compositional and functional diversity. This reinforces the use of spectral indices to assess the ecology of riparian plant communities and to further understand spatial patterns of ecosystem functioning across river networks. Given the seasonal variation in the patterns, we encourage the analysis of the whole phenological cycle instead of single‐date indices.