Hyperdominant Trees Reveal Savanna Vulnerability Under Climate Change

Abstract

The Cerrado biome, spanning ~2 million km2, is one of the most extensive and biodiverse tropical savannas, yet it is paradoxically dominated by only 30 hyperdominant tree species (~2% of all species, > 50% of all stems). However, their vulnerability to climate change and the effectiveness of current conservation efforts remain uncertain. By combining (i) species distribution models calibrated with edaphic-climatic predictors and occurrence data for hyperdominant Cerrado trees with (ii) functional-trait analyses related to ecological strategies (leaf economics, bark investment, ecophysiology, drought tolerance, regeneration, and dispersal), we identified the biomass production and reproductive traits that best predict species persistence under high-emission scenarios (RCP8.5). Currently, only 427,980 km2 (~17.4%) of the modeled potential environmental suitability falls within protected areas; under future climate scenarios, this protected suitable area is projected to decline by ~45.9% to 231,377 km2 (~18.1%). This loss, resulting from land conversion with the expansion of agricultural frontiers and fire events, highlights a mismatch in current regional conservation priorities, overlooking the needs of the Cerrado and its associated biodiversity. Given ongoing land-use change and that 58% of remaining native vegetation occurs on private lands, this mismatch highlights the urgent need to align conservation and agricultural policy. Functional trait analyses revealed a clear gradient from acquisitive “fast” to conservative “slow” strategies, reflecting trade-offs in water-use efficiency and biomass allocation, with leaf structural and reproductive traits best predicting species resilience to climate change. Our results indicate that a small subset of hyperdominant species not only forms current community structure but also signals the biome's resilience/vulnerability to climate change. Conservation planning should prioritize identified climate refuges and be based on functional traits to buffer the loss of functional and structural integrity in one of the world's richest and most threatened savanna ecosystems.