Using Conventional Observations and Coyote UAS Data to Evaluate the Grell-Freitas Cumulus Scheme in the Hurricane Weather Research and Forecast System

The parameterization of cumulus convection in numerical weather prediction models plays an important role in modulating tropical cyclone structure, intensity, and evolution. The Hurricane Weather Research and Forecast system (HWRF), the National Oceanic and Atmospheric Administration’s operational model for tropical cyclone prediction, currently uses the scale-aware Simplified Arakawa-Schubert (SASAS) cumulus scheme. In this research, the impact of replacing SASAS with the Grell-Freitas (GF) scheme was studied. The GF scheme was selected because it is a state-of-the-art scheme that employs an ensemble approach to the representation of convection, using a collection of parameters and algorithms to represent convective triggers, vertical mass flux, and closures. It is also scale aware, which makes it suitable for use across a wide range of model horizontal grid spacings. The scheme is already used in the Rapid Refresh (RAP) that is run operationally by the National Centers for Environmental Prediction (NCEP).

To assess the suitability of GF for HWRF simulations of tropical cyclones (TCs), retrospective HWRF forecasts of selected TCs that occurred in the Atlantic and eastern North Pacific ocean basins from 2015–2017 were simulated using the GF scheme and compared to those produced by the operational HWRF configuration. Both traditional model verification metrics (i.e., tropical cyclone track and intensity) and a comparison between the modeled and observed air temperature, moisture, winds, and precipitation will be presented for each model configuration, with a focus on Hurricane Maria (2017). Maria was observed heavily by both crewed Hurricane Hunter aircraft and the ground-based Weather Service Radar–1988 Doppler (WSR-88D) in San Juan, Puerto Rico prior to its destruction by the eyewall of the storm. In addition, the Coyote Uncrewed Aircraft System (UAS) completed six flights in the boundary and surface layers of the storm, providing an unprecedented dataset of thermodynamic and kinematic variables at heights as low as 100 meters above the ocean surface. The combination of uncrewed and crewed aircraft observations and precipitation data from the WSR-88D make Maria a unique opportunity to conduct a model evaluation comparing the SASAS and GF cumulus schemes.