Climate change is already affecting the terrestrial water and carbon cycle in ways that remain yet uncertain. In this study, I investigate changes in rainfall variability, and its extremes, and how the coupled water/carbon dynamics in natural terrestrial ecosystems respond to altered rainfall. I do this first by assessing the dependency of rainfall extremes on surface air and dewpoint temperature, based on station data, a reanalysis dataset, and recently published continental-wide high-resolution convection-permitting climate model simulations. For the first time, I reveal the global scaling pattern of rainfall extremes with surface air and dewpoint temperature at fine spatiotemporal scales, and demonstrate the relevance of thermodynamical constrains over the mid- and high-latitudes, while also identifying hotspots of divergence from such expectations. I further assess convection-permitting model estimates of changes in rainfall extremes, their seasonality, and duration, and I discuss the physical link between changes in the statistical properties of rainfall extremes, thermodynamical expectations, and the differential responses of events of convective and non-convective nature. In turn, I stochastically extend convection-permitting model simulations of the current and future climate with a stochastic weather generator, and study impacts over 33 North American ecosystems, covering a wide range of climates and biomes, by employing a state-of-the-art terrestrial biosphere model. I fully disentangle changes in mean annual rainfall, rainfall seasonality, and changes in the fine-scale temporal structure of rainfall, and study their separate effects. To investigate more deeply the revealed irrelevance of changes in the fine-scale temporal structure of rainfall on ecosystem productivity, I conduct further numerical experiments, by manipulating rainfall following the "fewer but larger events" concept, which is commonly applied in field rainfall manipulation experiments, and discuss the implications of my findings. Finally, I study ecosystem impacts under total climatic change, and investigate the separate and joint responses of the coupled water/carbon dynamics to altered rainfall, elevated CO2, and increased temperature and atmospheric drought, focusing mostly on transpiration. For the first time, high-resolution convection-permitting model outputs are used to assess what robust high-resolution estimates of changes in the climatic forcing suggest about future ecosystem transpiration, and the terrestrial carbon sink.