Skip to main content
Dryad

Interspecific competition negates CO₂ benefits for most plant species

Raubenheimer, Sarah1

Published Feb 16, 2026 on Dryad. https://doi.org/10.5061/dryad.8sf7m0d2q

Data files

Feb 16, 2026 version files 469.75 KB

Click names to download individual files

Abstract

Elevated CO₂ (eCO₂) frequently increases plant growth and primary productivity, and is considered a driver of increased land carbon sinks. This evidence contributes to a widespread but surprisingly untested assumption in global change science and policy, and society at large, that eCO₂ consistently stimulates the growth of plant species within communities. However, species and community responses depend on species’ growth strategies and abiotic and biotic contexts, and among these, the role of competition remains underexplored. We therefore ask whether eCO2 enhances growth broadly across species, or primarily in a dominant few? To address this question of how species interactions are shaped by eCO₂ response and competitive ability, we analysed responses from 97 species in 20 grassland CO₂ enrichment experiments. Herein, we show that interspecific competition markedly reduced eCO₂ stimulation: while most species responded positively to eCO₂ in isolation, over half responded negatively in mixed communities. Despite this, community biomass was often and on average was increased under eCO₂, driven by dominant species contributions. eCO₂ did not enhance competition tolerance, and acquisitive traits—including fast growth and invasion success—conferred a consistent advantage under both CO₂ enrichment and competitive stress. Our findings suggest that in species mixtures, which represent the majority of natural ecosystems, competition causes both positive and negative growth responses to eCO2 to be ubiquitous and pervasive. We show that eCO₂ does not consistently reduce competitive vulnerability, but instead, trait-based filtering in a competitive context determines both which species benefit from eCO2. These results challenge assumptions built into ecosystem models; and incorporating these dynamics will improve forecasts of vegetation response to rising CO₂.