Improving Coyote UAS Tropical Cyclone Sampling Strategies in an OSSE

The Coyote is a small uncrewed aircraft system (UAS) deployed from the NOAA P-3 aircraft that continuously samples temperature, humidity, wind, and pressure over a flight duration of up to one hour. It has the potential to fill an important data gap in the boundary layer of tropical cyclones, where in-situ sampling is typically limited to point measurements from dropsondes and buoys due to the dangers associated with flying crewed aircraft below ~3 km AGL (Cione et al. 2016). To date, the Coyote has been successfully deployed twice in Hurricane Edouard (2014) and six times in Maria (2017), targeting the eyewall and inflow layer of the storm. Previous experiments with data from the Edouard missions showed positive impacts from assimilating Coyote observations in a vortex-scale data assimilation system. However, greater impact from the observations may be achieved by using different sampling strategies.

In this study, we investigate how to improve Coyote UAS sampling strategies for model initialization by performing an observing system simulation experiment (OSSE), where observations are simulated from the Nolan et al. (2013) hurricane nature run and assimilated using NOAA’s HWRF and HEDAS modeling and data assimilation systems. These results will be compared against observing system experiments (OSEs) with real Coyote data. Considerations for future expansion alternatives will also be presented, including longer battery life that allows for longer flight times and advances in current sensors that allow for more frequent in-situ sampling of the vortex.