Abstract: In the Ensemble Kalman Filter (EnKF) data assimilationprediction system, most of computation time is spent in the prediction runs of the members. A limited or smaller ensemble size does reduce the computational cost, but an excessively small ensemble size usually leads to filter divergence, especially when there are model errors. In order to improve the time efficiency of the EnKF data assimilationprediction system and prevent it against filter divergence, a time-xpanded sampling approach for the EnKF based on the WRF (Weather Research and Forecasting) model is used to assimilate simulated sounding data and investigate the assimilation effect of the approach in a mesoscale model. The approach samples a series of perturbed state vectors from the Nb member prediction runs not at the analysis time (as the conventional approach does) but also at equally separated time levels (time interval is Δt) before and after the analysis time with M times. All the above sampled state vectors are used to construct the ensemble and compute the background covariance for the analysis, so the ensemble size is increased from Nb to Nb+2M×Nb=(1+2M)×Nb without increasing the number of prediction runs (is still Nb). This reduces the computational cost. A series of experiments are conducted to investigate the impact of Δt (the time interval of timeexpanded sampling) and M (the maximum sampling times) on the analysis. The results show that if Δtand M are properly selected, the time-expanded sampling approach achieves the similar effect to that from the conventional approach with an ensemble size of (1+2M)×Nb, but the number of prediction runs is greatly reduced.