Aircraft observations are known to be the most important type of data assimilated by short-range Numerical Weather Prediction (NWP) models for predicting winds and temperatures (James et al. 2020), yet current aircraft data systems (such as MDCRS, a.k.a. ACARS) only obtain observations from a small percentage of large commercial aircraft. Over 95% of commercial jets plus many smaller aircraft are equipped with Mode Selective (S) Enhanced Surveillance (EHS) transponders which, if they are appropriately interrogated, can be used as a source of wind and temperature estimates. Six Portable Aircraft Derived Weather Observation Systems (PADWOSs) have been deployed across four states in the southern United States via a joint program between NOAA and MIT Lincoln Laboratory. These systems can interrogate Mode S EHS equipped aircraft that fly within 60 nautical miles of a PADWOS and provide temperature and wind estimates in real-time. We assimilate data from PADWOSs into the Rapid Refresh Forecast System (RRFS), which is a convection-allowing (3-km grid spacing) ensemble-based data assimilation and forecasting system. We conduct retrospective RRFS simulations to quantify the impact of assimilating PADWOS data on model forecast accuracy. We explore numerous assimilation strategies and quantify their impacts relative to a control run as well as simulations with and without other types of aircraft data.