High-resolution regional reanalysis is one of the most powerful tool to study local and regional high-impact weather events, climate extremes and climate change impact. In this study, two high-resolution Chinese Regional Reanalysis (CNRR) datasets with a resolution of 18km over 1998-2012, which are produced using the Gridpoint Statistical Interpolation (GSI) data assimilation system and spectral nudging (SN) technique, were assessed. The reliability of the surface and upper air variables of the CNRR datasets was evaluated by comparing with in-situ observations and the European Centre for Medium-Range Forecasts (ECMWF) ERA-Interim (ERAIN) global reanalysis dataset. The results show that the CNRR can provide more accurate near-surface variables than the driving ERAIN global reanalysis. However, special care should be taken when using the CNRR precipitation dataset, especially for heavy rainfall cases. CNRR datasets are also able to generate high-quality upper atmospheric products especially in CNRR-GSI experiment, which assimilates long time series of local observations. By using the three-dimensional variational data assimilation (3D-Var) method, CNRR-GSI outperforms the CNRR-SN and ERAIN. With the increase of the computing resources, the potential opportunities for improving CNRR can be expected by applying more advanced methods.