Initial evaluation of the hourly winds from WACCM-NEDAS data assimilation using multiple datasets

Characterizing hourly wind variability in the mesosphere and lower thermosphere (MLT) remains challenging because direct observations are sparse and observational constraints on gridded products weaken rapidly toward the mesopause. Here we present an initial evaluation of January–March 2019 winds from a whole-atmosphere analysis produced by coupling the Whole Atmosphere Community Climate Model version 6 (WACCM6) with the Next-generation Ensemble Data Assimilation System (NEDAS) using a two-stage framework. Below ~80 km, we perform cross-product consistency checks against the European Centre for Medium-Range Weather Forecasts fifth-generation atmospheric reanalysis (ERA5), the Modern-Era Retrospective analysis for Research and Applications, version 2 (MERRA-2), the specified-dynamics WACCM (SD-WACCM), and the Horizontal Wind Model 2014 (HWM14) on a common grid. Above 76 km, hourly winds are independently evaluated against nonassimilated meteor radar observations over Wuhan (30.6°N, 114.4°E). In the upper stratosphere and lower mesosphere, WACCM-NEDAS reproduces the dominant large-scale zonal-mean circulation and remains broadly consistent with ERA5 and MERRA-2, with the most coherent agreement against MERRA-2 through much of the 60- to 80-km layer. In the MLT, WACCM-NEDAS is closer to the meteor radar winds than is SD-WACCM, with full-sample correlations of 0.55 for the zonal wind and 0.67 for the meridional wind and height-resolved correlations reaching 0.7–0.8 in favorable layers. At a representative 90-km level, wavelet analysis further indicates that WACCM-NEDAS retains a broader short-period variability spectrum than does SD-WACCM and avoids the overly smoothed near-diurnal response of HWM14. A supplementary 2- to 6-d wavelet analysis during the 2018–2019 sudden stratospheric warming interval further shows that WACCM-NEDAS captures a 4- to 5-d enhancement consistent with the meteor radar spectra. These results indicate meaningful added value of WACCM-NEDAS for hourly MLT wind studies, although broader validation across additional stations, seasons, and dynamical conditions is still required.