Floral traits predict nectar rewards across diverse plant communities

Floral nectar is a key energy resource for many animals, yet direct measurement of nectar rewards is time consuming and difficult to apply across taxonomically diverse plant communities. We tested whether certain floral traits can predict nectar rewards across a broad taxonomic and morphological range of flowering plants. We analysed relationships between floral traits and nectar rewards in 156 species from 42 families. Corolla tube length, corolla base width and corolla shape were recorded alongside nectar volume, sugar concentration and total sugar content, and linear mixed-effects models were used to quantify floral trait nectar relationships in tubular and non tubular flowers. Predictive performance was evaluated using observation- and species-level cross-validation, and phylogenetic signal in nectar volume was assessed using Blomberg's K and Pagel's {lambda}. Nectar volume increased strongly with floral dimensions in both tubular and non-tubular flowers, while sugar content was closely associated with nectar volume, with an approximately proportional relationship on the log scale. These relationships support the use of floral traits to predict sugar content, with predictive performance retained for species not included in model fitting (species-level cross-validation R^2 = 0.43 for tubular and 0.51 for non-tubular flowers). Nectar volume also showed weak but significant phylogenetic signal. Overall, our findings demonstrate that simple floral traits can provide useful predictors of potential nectar rewards across a taxonomically and morphologically diverse range of flowering plants. Such an approach opens the way for future studies to combine trait-based predictions with information on floral abundance, flowering phenology to investigate how nectar resources vary across landscapes in space and time, with potential applications in pollination ecology and the study of nectar feeding insects. Keywords: corolla morphology, floral traits, nectar sugar content, nectar volume, trait based prediction.