The Impacts of Assimilating Various Densities of Uncrewed Aircraft System Observations on Regional NWP Forecasts in an OSSE

Abstract Uncrewed aircraft systems (UAS) have emerged as an option for increasing the number of routine observations within the in situ observational gap in the lower troposphere. Before deploying a nationwide network of UAS, however, it is necessary to determine what impact UAS observations will have on weather forecast model accuracy and assess the relative benefits of various UAS networks. Our goal is to help address this knowledge gap by examining the impact of assimilating varying densities of UAS observations on Rapid Refresh Forecast System (RRFS) forecasts. To do this, an observing system simulation experiment (OSSE) is used that consists of two week-long nature runs over the contiguous United States. Five different networks in which UAS execute hourly vertical profiles up to 2 km AGL are examined, with the spacing between UAS sites varying between 300 and 35 km. Results show positive impacts from assimilating UAS, with observations from the 35-km UAS network reducing 6-hour root-mean-squared errors by over 15% in the lower atmosphere. It is also shown that the benefit per UAS in the bulk verification statistics decreases as more UAS are added to the network. Examining a low cloud ceiling case shows that UAS can improve cloud forecasts when there are minimal clouds at the analysis time owing to a better representation of above-ground moisture, though the UAS impact was minimal when using the coarsest UAS network. Altogether, these results suggest that UAS can improve RRFS forecasts and benefits can be obtained from less than a hundred UAS.