Development of a Hybrid En3DVar Data Assimilation System and Comparisons with 3DVar and EnKF for Radar Data Assimilation with Observing System Simulation Experiments

Abstract

A hybrid ensemble–3DVar (En3DVar) system is developed and compared with 3DVar, EnKF, “deterministic forecast” EnKF (DfEnKF), and pure En3DVar for assimilating radar data through perfect-model observing system simulation experiments (OSSEs). DfEnKF uses a deterministic forecast as the background and is therefore parallel to pure En3DVar. Different results are found between DfEnKF and pure En3DVar: 1) the serial versus global nature and 2) the variational minimization versus direct filter updating nature of the two algorithms are identified as the main causes for the differences. For 3DVar (EnKF/DfEnKF and En3DVar), optimal decorrelation scales (localization radii) for static (ensemble) background error covariances are obtained and used in hybrid En3DVar. The sensitivity of hybrid En3DVar to covariance weights and ensemble size is examined. On average, when ensemble size is 20 or larger, a 5%–10% static covariance gives the best results, while for smaller ensembles, more static covariance is beneficial. Using an ensemble size of 40, EnKF and DfEnKF perform similarly, and both are better than pure and hybrid En3DVar overall. Using 5% static error covariance, hybrid En3DVar outperforms pure En3DVar for most state variables but underperforms for hydrometeor variables, and the improvement (degradation) is most notable for water vapor mixing ratio qυ (snow mixing ratio qs). Overall, EnKF/DfEnKF performs the best, 3DVar performs the worst, and static covariance only helps slightly via hybrid En3DVar.

A 3D ensemble variational data assimilation scheme for the limited‐area AROME model: Formulation and preliminary results

“…Increasing the size of the ensemble (e.g. by using time‐lagged ensembles), combined with an increase of the weight of the ensemble part, seems particularly beneficial as is shown at global (Lorenc, ), regional (Gustafsson et al , ), and convective (Kong et al , ) scales. More weight could also be given to the climatology for the higher levels (e.g.…”

Techniques and challenges in the assimilation of atmospheric water observations for numerical weather prediction towards convective scales

“…The increment is spread out due to the B-matrix, and is horizontally and vertically symmetric due to the way that spatial covariances are modelled. The part of the increment below the freezing level (around the level of the observation) is unrealistic -ideally the increments would match those of Figure 3a which is the increment required to give the true hail in this simulation (for details, see Kong et al (2018)). This kind of problem justifies development of Ens-based DA systems.…”

“…Used with permission. Taken fromKong et al (2018) [Colour figure can be viewed at wileyonlinelibrary.com].…”

“…En3D-Var is the version of EnVar equivalent to 3D-Var but with ensemble-derived covariances. 5 DfEnKF is an implementation of the EnKF used byKong et al (2018).…”

Impacts of Humidity Adjustment Through Radar Data Assimilation Using Cloud Analysis on the Analysis and Prediction of a Squall Line in Southern China

“…To effectively use the mesoscale and convective‐scale information observed by Doppler radar, many efforts had been made based on different DA methods, including the three‐dimensional variational (3‐DVar) and cloud analysis frameworks (Gao et al, ; Hu, Xue, & Brewster, ; Hu, Xue, Gao, & Brewster, ; Liu et al, ), the four‐dimensional variational (4‐DVar) framework (Sun & Crook, ), the ensemble Kalman filter (EnKF) (Houtekamer & Zhang, ; Jung et al, ; Tong & Xue, ; Xue et al, ), and hybrid methods (Gao & Stensrud, ; Kong et al, ; Li et al, ).…”