Summary: Climate change is increasing global temperatures and the frequency and severity of droughts in many regions. These anthropogenic stresses pose a significant threat to plant performance and crop production. The plant‐associated microbiome modulates the impacts of biotic and abiotic stresses on plant fitness. However, climate change‐induced alteration in composition and activities of plant microbiomes can affect host functions. Here, we highlight recent advancements in our understanding of the impact of climate change (warming and drought) on plant–microbiome interactions and on their ecological functions from genome to ecosystem scales. We identify knowledge gaps, propose new concepts and make recommendations for future research directions. It is proposed that in the short term (years to decades), the adaptation of plants to climate change is mainly driven by the plant microbiome, whereas in the long term (century to millennia), the adaptation of plants will be driven equally by eco‐evolutionary interactions between the plant microbiome and its host. A better understanding of the response of the plant and its microbiome interactions to climate change and the ways in which microbiomes can mitigate the negative impacts will better inform predictions of climate change impacts on primary productivity and aid in developing management and policy tools to improve the resilience of plant systems.

A lack of empirical evidence for the microbial regulation of ecosystem processes, including carbon (C) degradation, hinders our ability to develop a framework to directly incorporate the genetic composition of microbial communities in the enzyme-driven Earth system models. Herein we evaluated the linkage between microbial functional genes and extracellular enzyme activity in soil samples collected across three geographical regions of Australia. We found a strong relationship between different functional genes and their corresponding enzyme activities. This relationship was maintained after considering microbial community structure, total C and soil pH using structural equation modelling. Results showed that the variations in the activity of enzymes involved in C degradation were predicted by the functional gene abundance of the soil microbial community (R2>0.90 in all cases). Our findings provide a strong framework for improved predictions on soil C dynamics that could be achieved by adopting a gene-centric approach incorporating the abundance of functional genes into process models.