Back

Effects of temperature gradient on flower and fruit traits: a meta analysis

Nevo, O.; Chronopoulou, E. L.; Azeroth, A.; Rakosy, D.; Onstein, R. E.; Knight, T. M.; Kuppler, J.

2026-08-06 ecology
10.64898/2026.08.06.743200 bioRxiv
Show abstract

Pollination and seed dispersal by animals are key drivers of terrestrial biodiversity and ecosystem functioning. Effective mutualistic interactions rely heavily on temporal and functional- trait matching between plants and animals. While global warming is known to induce shifts in plant traits, the extent and direction in which higher temperatures may systematically alter flower and fruit traits across species in natural habitats remains poorly understood. Using elevation as a proxy for temperature across natural populations, we conducted a meta-analysis evaluating 21 quantitative functional traits (15 floral, 6 fruit) across 82 studies and 161 species. Standardized mixed-effects linear regression models revealed widespread, systemic responses to elevational temperature gradients in both reproductive structures. In flowers, higher temperatures were systematically associated with changes in petal and sepal width and length (and hence morphology), longevity, nectar volume, number of flowers, and inflorescence length. In fruits, elevation was associated with changes in vitamin C content, crop size, weight and width. Taken together, these results demonstrate that warming temperatures exert widespread, multi-axis effects on the morphology, availability, timing, and nutritional quality of both flowers and fleshy fruits. Given that flower and fruit traits are developmentally linked and co-determine animal visitor dynamics, these temperature-driven phenotypic shifts are likely to propagate cascading disruptions throughout plant-pollinator and plant-frugivore interaction networks under continued climate change.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.