In arctic tundra, shrubs can significantly modify the distribution and physical characteristics of snow, influencing the exchanges of energy and moisture between terrestrial ecosystems and the atmosphere from winter into the growing season. These interactions were studied using a spatially distributed, physically based modelling system that represents key components of the land–atmosphere system. Simulations were run for 4 years, over a 4-km2 tundra domain located in arctic Alaska. A shrub increase was simulated by replacing the observed moist-tundra and wet-tundra vegetation classes with shrub-tundra; a procedure that modified 77% of the simulation domain. The remaining 23% of the domain, primarily ridge tops, was left as the observed dry-tundra vegetation class. The shrub enhancement increased the averaged snow depth of the domain by 14%, decreased blowing-snow sublimation fluxes by 68%, and increased the snowcover's thermal resistance by 15%. The shrub increase also caused significant changes in snow-depth distribution patterns; the shrub-enhanced areas had deeper snow, and the non-modified areas had less snow. This snow-distribution change influenced the timing and magnitude of all surface energy-balance components during snowmelt. The modified snow distributions also affected meltwater fluxes, leading to greater meltwater production late in the melt season. For a region with an annual snow-free period of approximately 90 days, the snow-covered period decreased by 11 days on the ridges and increased by 5 days in the shrub-enhanced areas. Arctic shrub increases impact the spatial coupling of climatically important snow, energy and moisture interactions by producing changes in both shrub-enhanced and non-modified areas. In addition, the temporal coupling of the climate system was modified when additional moisture held within the snowcover, because of less winter sublimation, was released as snowmelt in the spring.