Differences in adult survival drive divergent demographic responses to warming on the Tibetan Plateau.

A central question in biodiversity conservation is whether species will maintain viable populations under climate warming. Assessing species viability under climate warming requires demographic studies integrating vital rate responses to long-term warming throughout species' life cycles. However, studies of this nature are rare. Our integral projection models (IPMs), parameterized with demographic data, show differing responses of two functionally similar co-occurring species, Elymus nutans Griseb. and Helictotrichon tibeticum (Roshev.) Holub, to 10 years of in situ active warming by 2°C. Our IPMs estimated that the life expectancy is higher in H. tibeticum (6.7 years) than that in E. nutans (4.5 years) under ambient conditions, and the difference is larger under warmed conditions. We found that while warming decreased individual-level growth in both species, H. tibeticum, which has a longer life expectancy, compensated with increased survival, and thereby increased projected population-level growth under warming. Contrastingly, E. nutans, which has a shorter life expectancy, is projected to have decreased population-level performance. Furthermore, our elasticity analyses show that survival is the most important vital rate for population viability in both species under both ambient and warmed conditions. Moreover, our retrospective life table response experiment (LTRE) analysis reveals that the contrasting fates of the two species under warming mainly arise from the different responses of adult survival, which is significantly promoted in H. tibeticum but slightly reduced in E. nutans. Individual shrinkage occurred 1.6 fold more frequently under warming than ambient conditions for both species and made considerable negative contributions to their population growth rates in warmed plots. However, such negative effects are offset in H. tibeticum (but not E. nutans) by the positive contribution to population growth rate of the associated increased survival. Our results illustrate that the responses to climate warming may vary considerably between similar co-occurring species, and species with a demographically compensatory strategy may avoid population collapse. Furthermore, our study demonstrates the potential of using life-history traits to predict species' viability when facing warming, so as to inform biodiversity conservation under climate change.