Mesoscale convective systems (MCSs) are organized convective weather systems that cover an area of several hundred kilometers and last for several hours or more. To improve the observation and prediction of MCSs, a dense network of observing systems is required. The surface Mesonet provides better coverage of surface observations than conventional meteorological networks. By assimilating surface meteorological Mesonet observations into numerical weather prediction (NWP) models, more accurate initial conditions can be obtained. Ground-based radar and lidar data offer valuable information about the structure and evolution of MCSs. Radar data, such as reflectivity and Doppler velocity, provide insights into precipitation patterns, storm intensity, and wind fields within the storms. Lidar data, which uses laser beams to measure atmospheric properties, can provide vertical wind profiles. Assimilating radar and lidar data into NWP models can enhance the representation of convective processes within MCSs, resulting in improved predictions of their track, intensity, and timing. Data assimilation involves incorporating these observational data into numerical models using variational or ensemble-based assimilation methods. These methods aim to find the best fit between the observations and the model's simulated state, effectively blending the observations with the model's background state. The assimilation process enables a more accurate representation of the initial atmospheric conditions and better captures the complex dynamics and microphysics associated with these convective systems, ultimately enhancing the skill and reliability of MCS forecasts. This chapter utilizes MCS cases over the Southern Great Plains of the United States to illustrate the significance of assimilating surface Mesonet meteorological observations, radar reflectivity and radial velocity, and lidar profiling data into the mesoscale Weather Research and Forecasting (WRF) model.