Improvements in Convective Scale Ensemble Forecasts from the Real-Time Assimilation of Ground-Based Thermodynamic Profilers

Ground-based remote sensing thermodynamic profilers have been repeatedly demonstrated to have value for research into atmospheric processes, and the operational benefits of such systems for near-real time monitoring of environmental changes and situational awareness are becoming more well-known. The impact of these profiles on numerical weather prediction (NWP), however, has been relatively unexplored. While some work has been done showing the impact of real or simulated profiler observations in NWP for retrospective forecasts (or hindcasts), little has been done to illustrate how a network of profilers can benefit real-time operational NWP needs.

In the present work, we demonstrate the implementation of thermodynamic profiles retrieved from a network of Atmospheric Emitted Radiance Interferometers (AERIs) and the utility that these observations have on a convective-scale NWP ensemble. AERI is a ground-based hyperspectral infrared radiometer capable of retrieving temperature and moisture profiles in the lower troposphere with a temporal resolution of 10 min or better. A network of five AERIs was erected in 2017 throughout the Atmospheric Radiation Measurement (ARM) program’s Southern Great Plains (SGP) research facility. During the 2018 NOAA Hazardous Weather and Hydrometeorology Testbed experiments, observations from these instruments were collected, processed into retrievals, and assimilated (with less than 15 min latency) into the National Severe Storms Laboratory (NSSL) Experimental Warn-on-forecast System for Ensembles (NEWS-e), a WRF-based 36-member ensemble with nested grids as fine as 3 km.

AERI was a successful member of the suite of observations assimilated by NEWS-e during the 2018 experiment season, and examples of this will be presented. Analyses of hindcasts performed after the spring experiments show the positive impact that these observations have on forecast skill. Plans for the 2019 experiments, in which improved retrievals that include realtime cloud-base height observations, will also be discussed.