The McMurdo Dry Valleys of Antarctica are an abiotically driven ecosystem characterized by having a very simple trophic structure dominated by microbial communities, whose diversity is shaped by extreme abiotic gradients, particularly extreme aridity and oligotrophy. Regional isolation and dispersal limitations have concurrently led to the emergence of heterogeneous microbial communities with highly localized dominant taxa across the region. These taxa were selected based on specialized genetic and physiological adaptations accumulated during long-term isolation, which conferred an advantage to endure the physical and chemical stress. Models predict that over the coming decades, climate change will trigger hydrological changes in the system with potential consequences for its microbial communities and, subsequently, ecosystem-level processes. The capacity of the Antarctic microbiome to absorb change while maintaining its structural and/or functional attributes will determine the extent to which predicted environmental changes will threaten the system's stability. This research starts by developing and validating a space-for-time sampling approach using variations in geochemical factors that follow alterations in water availability as time progresses and to which biological communities respond. This approach was replicated across the six major lakes in the Wright and Taylor valleys, and builds on previous examples of environmental gradients, which used arbitrary distance-based metrics as sampling design, incorporating significant yet uncharacterized in situ geochemical variability. The approach developed here enabled the acquisition of a comprehensive dataset that predicts, with confidence, that future hydrological changes will significantly alter the composition and diversity of microbial communities historically adapted to arid and oligotrophic conditions. The latter will result in significant changes in the metabolic activity of pathways associated with carbon, nitrogen, phosphorous and sulphur cycles, with an increase in functional diversity and activity as the system becomes wetter. This work further provides first time evidence that carbon fixation via atmospheric chemosynthesis is the primary active pathway for carbon acquisition under extreme aridity in polar deserts, being replaced by photosynthetic carbon fixation with prolonged exposure to wetness. To complement predictions made in situ using a space-for-time approach, this research incorporated temporal observations using manipulative experiments, performed in a Polar Desert Environmental Chamber (PDEC), to test the sensitivity and resilience of microbial communities to short-term wetting disturbances. It demonstrated that co-existing microbial taxa respond asynchronously during wetting and drying periods, which indicates dry soil communities comprise co-existing taxa with preferences for different environmental conditions. It also experimentally showed, for the first time, the capacity of microbial communities from this region to recover from short-term wetting events associated with the ability of dry-adapted taxa, mostly affiliated with Actinobacteriota and Acidobacteria phyla, to persist during the wetting period. Through the incorporation of large-scale spatial transects in conjunction with manipulative experiments, this research delivered a fundamental evidence-based anticipation of the compositional and functional aspects that are likely to change in the McMurdo Dry Valleys ecosystem in response to alterations in water dynamics as well as on the microbial attributes that enhance the ecosystem's functional resilience to climate change.