Global climate model simulations are commonly used to make future climate projections both directly and indirectly (by providing boundary conditions to regional models). Bias correction methods are often applied to climate model output based on the assumption that biases are constant in time, which may or may not be valid. Past studies have tested this assumption regionally using a small subset of climate models and found that precipitation and temperature biases typically varied in time and by region, which may have been overlooked in impact studies. Model biases can introduce substantial uncertainty into future climate projections, motivating the need to understand any relationships between model biases and future projections. In this study, we investigated the impact of temperature and precipitation biases on future climate projections over the central United States. We utilized global climate models that participated in phase six of the Coupled Model Intercomparison Project (CMIP6) for which data was available. We plan to quantify model biases by comparing observations and simulated seasonal climatologies over two twenty-year historical periods (1961-1980 and 1981-2000). By assessing the model biases over two time periods, we can determine to what extent the biases are stationary in time. In addition, we plan to investigate any relationships between the historical climate model biases and future climate projections, including dependence on season and variable type (precipitation and temperature). This research will help to inform the selection of global models used for impact studies and downscaling and to better understand how the occurrence of model biases may introduce uncertainty into future climate projections.