There is strong empirical evidence on the importance of the spatial pattern of vegetation in dryland hydrologic and geomorphologic dynamics. However, changes in vegetation cover and spatial pattern are often linked, making it difficult to disentangle and assess their independent hydro-geomorphologic roles. We used synthetic sponges placed on the soil surface to mimic the aboveground structure of vegetation patches, and manipulated patch cover and pattern as well as the sink capacity of the patches on a set of 24 (2 × 1 m) runoff plots. Combining natural-rainfall and simulated-rainfall experiments, we aimed to test that (1) both vegetation cover and pattern independently control runoff and sediment yield; (2) for any given cover, coarsening the vegetation pattern entails increasing runoff and sediment yield; and (3) pattern effect is mostly exerted by modulating the source-sink dynamics of the system. We found that increasing either patch cover or patch density decreased runoff and sediment yields from natural rainfalls, yet the effect of patch density largely disappeared when the effect of the co-varying patch cover was removed. Simulated-rainfall experiments on plots with equal medium-low patch cover showed however that coarser patterns (lower patch density; higher patch size) increased runoff coefficients and reduced time to runoff as compared with finer patterns. The effect of patch density was particularly clear when the sink function of vegetation patches was also mimicked. Rainfall interception and direct soil protection proved to be critical mechanisms underlying the effects of patch cover, yet they barely contributed to the effects of patch pattern. The control of overland flow by patch pattern was exerted through changes in the level of runoff disruption. However, physical obstructions to runoff hardly reduced runoff unless coupled to mimicked soil sinks. Overall this work demonstrates the independent effects of patch cover and pattern on the hydro-geomorphologic functioning of patchy landscapes, with patch cover being the primary hydrologic control factor and patch pattern exhibiting its full potential for low and medium low patch cover values. Our findings provide useful information for modelling and understanding dryland vegetation dynamics, and for designing management and restoration measures that take into account the critical role played by source-sink dynamics and hydrological connectivity in dryland landscapes.