Mesoscale Convective Systems (MCSs) are a major source of the warm-season precipitation over the central United States, particularly for extreme rainfall events. MCSs exhibit characteristic mesoscale circulations, but their top-heavy latent heating also influences circulations at larger scales. This dynamical upscale effect implies that the lack of MCSs in simulations not only leads to bias in precipitation but also in large-scale circulations, both locally and in remote locations. To quantify the dynamical upscale effect of MCSs, a series of hindcasts are conducted using the Model for Prediction Across Scales (MPAS) with grid spacings ranging from 50 km to convection permitting 4 km over the continental United States. With local grid refinement in the MPAS global variable resolution model, the dynamical influence of MCSs simulated in the United States can upscale locally through the generation of potential vorticity (PV) and propagate to the coarse-resolution domain. MCSs in hindcasts at different grid resolutions will be identified and tracked using a newly developed FLEXible object TRacKeR (FLEXTRKR) algorithm. The PV coinciding with a tracked MCS in the observation or hindcast defines a PV anomaly associated with the MCS. By inverting the PV anomaly, we will determine the temperature and wind fields based on assumed balanced states, such as quasi-geostrophy, to quantify the upscale effect of the MCS. Comparison of the upscale effect from the observed and simulated MCSs will shed light on biases in large-scale circulation that result from model biases in simulating MCSs across a range of model resolution.