AbstractA transition from a predominantly offshore to an onshore rainfall phase over the west coast of India was simulated using three one‐way nested domains with 12, 4, and 1.33 km horizontal grid spacing in the Weather Research and Forecasting model. The mechanism of offshore–onshore rainfall oscillation and the orographic effects of the Western Ghats are studied. A convective parametrization scheme was employed only in the 12 km domain. A trough extending offshore from the west coast facilitates offshore rainfall. This trough is absent during the onshore phase, and rainfall occurs over the coast mainly via orographic uplift by the Western Ghats. The model overestimates rainfall over the Western Ghats at all resolutions as it consistently underestimates the boundary‐layer stratification along the coast. Weaker stratification weakens the blocking effect of the Western Ghats, resulting in anomalous deep convection and rainfall over its windward slopes. The 4 and 1.33 km domains simulate the offshore‐to‐onshore transition of rainfall but fail to capture a sufficient contrast in rainfall between land and sea compared with observations. The 12 km domain produces light rainfall, anchored along the coast, throughout the simulation period and, hence, gravely underestimates the offshore rainfall. The offshore rainfall persisted in the 4 and 1.33 km domains in a sensitivity experiment in which the Western Ghats were flattened. This suggests that orographic effects do not significantly influence offshore rainfall. In another experiment, the convective parametrization scheme in the 12 km domain was turned off. This experiment simulated the offshore and onshore rainfall phases correctly to some extent but the rainfall intensity was unrealistically high. Thus, a model with a horizontal grid spacing of , in which convection evolves explicitly, is desired for simulating the west‐coast rainfall variations. However, improvements in the representation of boundary‐layer processes are needed to capture the land–sea contrast.