Plant Adaptation in a Changing Climate: From Molecular Mechanisms to Ecosystem Resilience

Climate change exposes plants to combinations of heat, drought, altered nutrient supply, hypoxia, soil physical constraints and disturbance regimes whose effects are expressed first through heterogeneous cellular responses but ultimately matter at organismal, population, community and ecosystem scales. Single-cell and spatial multi-omics now resolve cell identity, transcription, chromatin accessibility and metabolite distributions with unprecedented anatomical specificity. This review critically evaluates whether these technologies can move beyond mechanistic description to improve prediction of plant adaptation and ecosystem resilience. Literature published from 1 January 2017 to 15 July 2026 was synthesised, with earlier foundational studies retained where necessary. The evidence is strongest for identifying cell-type-specific stress responses, regulatory states and inter-tissue signalling, including differentiated heat responses in roots, spatially distinct drought responses in leaves, chromatin-expression coupling under osmotic stress, and root responses to soil compaction and dryland environments. Evidence becomes progressively weaker across scales. Most single-cell experiments use limited genotypes, organs, developmental stages and controlled environments; inferred trajectories and regulatory links are often not tested as causal determinants of fitness; and direct links between molecular cell states, heritable local adaptation, community reassembly and ecosystem resistance or recovery remain rare. Landscape genomics, functional-trait ecology, remote sensing and ecological forecasting provide complementary frameworks for scaling, but they usually operate independently of cellular multi-omics. We therefore propose an evidence-calibrated multiscale framework in which molecular features are treated as candidate predictors only after perturbational validation, genotype-by-environment testing, field replication and explicit cross-scale uncertainty propagation. Single-cell and spatial multi-omics can materially strengthen climate-resilience research, but their greatest near-term contribution is likely to be mechanistic feature discovery and causal refinement rather than stand-alone ecosystem prediction. Reliable forecasting will require coordinated experiments that connect regulatory variation to whole-plant performance, population fitness, community functional composition and repeated measures of ecosystem resistance and recovery.