Assimilating Doppler radar data into cloud models for better weather predictions is challenging. This study shows that using the ensemble Kalman filter (EnKF) can help estimate the relationships between observed variables and the weather state accurately. By assimilating radar observations from a simulated supercell, the EnKF produced precise analyses of wind, temperature, moisture, and condensate after just 30 minutes of data. The EnKF's ability to track the reference solution is not solely due to stable system dynamics, but also relies on covariances between different variables and the initialization of ensemble members. While promising, there are still important issues to address, such as uncertainty in environmental data and forecast model errors.