Abstract: Significant uncertainties exist in the ability of numerical models to accurately predict heavy rainfall events during the pre-summer rainy season in southern China’ warm sector. A notable instance occurred on May 29–30, 2020, when heavy rain struck Guangdong Province’s coastline, but all operational numerical models at the time failed to predict it. Consequently, eight experiments were conducted to assess the influence of incorporating surface-intensive observations into simulations through a technique known as nudging. The study found that the control experiment (EXP1), which nudged all surface meteorological elements, successfully replicated the development of linear convection and the spatial and temporal evolution of heavy precipitation along the Guangdong coastline. Sensitivity experiments highlighted that nudging surface water vapor was crucial for convection initiation. This was primarily attributed to a swift increase in relative humidity from 80% to near saturation (99%) within 3 hours at low levels, accompanied by a marked reduction in convective inhibition (CIN), lifting condensation level (LCL), and level of free convection (LFC). Nudging surface temperature enhanced thermal buoyancy by amplifying potential temperature perturbations, thereby influencing convection initiation and organization. Without nudging the surface temperature, convection occurred later and was less organized. Nudging surface wind helps correct the near-surface southwesterly wind direction, bringing convection evolution and rainfall into better alignment with observations. Six additional experiments examined the effect of the nudging duration on simulations. The results suggest that nudging all surface meteorological elements for 6 hours produced results similar to the control run, effectively modeling heavy rainfall along the coastline. Although water vapor surged rapidly within the first 3 hours, maintaining it for another 3 hours facilitated rapid convection development. Thus, incorporating surface-intensive observations through nudging during the initial 6 hours of model integration can enhance the numerical prediction of southern China’s warm sector.