Comprehensive sensitivity analysis of the WRF model for meteorological simulations in the Arctic.

Accurate meteorological simulations are important in response to the rapid climate change and insufficient meteorological observations in the Arctic. In this study, we assessed the performance of the Weather Research and Forecasting (WRF) model in simulating meteorological parameters at two Arctic stations (Barrow and Summit) in April 2019, by using measurements and statistical parameters. Sensitivity tests for different planetary boundary layer (PBL) schemes, four-dimensional data assimilation (FDDA) and sea surface temperature (SST) were also performed to reveal their impacts on the accuracy of model simulations. The results demonstrated that the WRF model performs the best in predicting the surface pressure but the worst in predicting the winds at these two stations. The sensitivity tests showed that among the four tested PBL schemes (ACM2, MYJ, BL and YSU), the model equipping with the MYJ scheme behaves the best in predicting the meteorological parameters especially the winds at these two Arctic stations. Applying FDDA nudging methods can also significantly increase the accuracy of simulations. In addition, we found that updating a time-varying SST in the model may bring a two-sided influence on meteorological simulations in the Arctic, especially at coastal stations. • MYJ scheme was found to perform the best in predicting the winds at the two Arctic stations. • Applying FDDA nudging methods decrease the deviations in simulations of meteorological parameters, especially the winds. • Updating the SST exerts a stronger impact on simulations at the coastal station than at the inland station. [ABSTRACT FROM AUTHOR]