Deep convection is associated with a number of atmospheric hazards (e.g., hail, wind gusts, torrential rain) and is particularly frequent in the Southern Great Plains of the United States. On one hand, organized deep convection systems with extreme precipitation and damaging wind pose a profound threat to the urban areas with high population density and high-value assets; On the other hand, the urban areas with unique surface energy balance, atmospheric stability, and flow field may impact the deep convections locally and downstream. In this study, we aim to investigate the impact of the Dallas-Fort Worth metroplex on characteristic deep convection systems. The study design uses a nested approach wherein long-term observational data records of precipitation, hail, cloud microphysical parameters, and cloud top heights are analyzed to generate an observationally derived assessment of the impact of the conurbation. Then we perform an ensemble of simulations at a cloud-resolving scale for selected deep convection cases from different seasons using the Weather Research and Forecasting (WRF) model. We evaluate the simulations against measurements from the Automated Surface Observing System and both ground- and satellite-based remote sensing products. The optimal model configuration is then used in a sensitivity experiment where the events are re-simulated with (i) the urban surface removed and (ii) the roughness length of the urban surface reduced. Output from the three simulations of each event will then be compared to determine whether the urban surface has a detectable and repeatable impact on the convective systems and whether the impact is due primarily through the changed surface energy balance or the enhanced convergence due to the surface roughness.