Back

Unexpected global biodiversity responses to climate change arise from complex adaptation, seascape, and dispersal dynamics

Keggin, T.; Skeels, A.; Hagen, O.; Melian, C.; Waldock, C.

2026-03-05 ecology
10.64898/2026.03.03.707980 bioRxiv
Show abstract

Global marine biodiversity peaks in the tropics and is widely projected to shift polewards under climate warming as species track thermal optima. These projections rely largely on phenomenological models that treat species responses as shifts in climate suitability and neglect fundamental eco-evolutionary dynamics. The consequences of these dynamics in fluctuating populations for global biodiversity patterns therefore remain unresolved. We addressed this deficit by integrating local adaptation, dispersal, and gene flow - mediated trait homogenisation into spatially explicit global simulations of >1,400 tropical reef fishes under climate change. Biodiversity responses departed systematically from simple poleward redistribution. Depending on the interaction between dispersal and adaptive capacity, latitudinal richness peaks widened or narrowed, shifted equatorward or poleward, or collapsed entirely. These alternative regimes emerged from threshold-dependent interactions between seascape connectivity, metapopulation trait homogenisation, and accumulating thermal stress. Critically, greater dispersal enhanced persistence only below seascape-driven connectivity thresholds; beyond which homogenisation of traits across thermally heterogeneous populations disrupted local adaptation - particularly at metapopulation peripheries - and increased extinction risk. Continuous demographic feedback further generated cumulative stress and extinction tipping dynamics absent from snapshot suitability models. Our results show that eco-evolutionary dynamics can invert or destabilise expected climate-driven redistribution patterns, indicating that projections excluding these mechanisms may systematically misestimate the magnitude and direction of global biodiversity change in a rapidly warming ocean.

Matching journals

The top 6 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.