Abstract. The East Asia Regional Reanalysis (EARR) system is developed based on the advanced hybrid gain data assimilation method (AdvHG) using the Weather Research and Forecasting (WRF) model and conventional observations. Based on EARR, the high-resolution regional reanalysis and reforecast fields are produced with 12 km horizontal resolution over East Asia for 2010–2019. The newly proposed AdvHG is based on the hybrid gain approach, weighting two different analyses for an optimal analysis. The AdvHG differs from the hybrid gain in that (1) E3DVAR is used instead of EnKF, (2) 6 h forecast of ERA5 is used to be more consistent with WRF, and (3) the preexisting, state-of-the-art reanalysis is used. Thus, the AdvHG can be regarded as an efficient approach for generating regional reanalysis datasets thanks to cost savings as well as the use of the state-of-the-art reanalysis. The upper-air variables of EARR are verified with those of ERA5 for January and July 2017 and the 10-year period 2010–2019. For upper-air variables, ERA5 outperforms EARR over 2 years, whereas EARR outperforms (shows comparable performance to) ERA-I and E3DVAR for January 2017 (July 2017). EARR represents precipitation better than ERA5 for January and July 2017. Therefore, although the uncertainties of upper-air variables of EARR need to be considered when analyzing them, the precipitation of EARR is more accurate than that of ERA5 for both seasons. The EARR data presented here can be downloaded from https://doi.org/10.7910/DVN/7P8MZT (Yang and Kim, 2021b) for data on pressure levels and https://doi.org/10.7910/DVN/Q07VRC (Yang and Kim, 2021c) for precipitation.

No abstract

Abstract. The East Asia Regional Reanalysis (EARR) system is developed based on the advanced hybrid gain data assimilation method (AdvHG) using Weather Research and Forecasting (WRF) model and conventional observations. Based on EARR, the high-resolution regional reanalysis and reforecast fields are produced with 12 km horizontal resolution over East Asia for 2010–2019. The newly proposed AdvHG is based on the hybrid gain approach, weighting two different analysis for an optimal analysis. The AdvHG is different from the hybrid gain in that 1) E3DVAR is used instead of EnKF, 2) 6 h forecast of ERA5 is used to be more consistent with WRF, and 3) the pre-existing, state-of-the-art reanalysis is used. Thus, the AdvHG can be regarded as an efficient approach to generate regional reanalysis dataset due to cost savings as well as the use of the state-of-the-art reanalysis. The upper air variables of EARR are verified with those of ERA5 for January and July 2017 and the two-year period of 2017–2018. For upper air variables, ERA5 outperforms EARR over two years, whereas EARR outperforms (shows comparable performance to) ERA-I and E3DVAR for January in 2017 (July in 2017). EARR better represents precipitation than ERA5 for January and July in 2017. Therefore, though the uncertainties of upper air variables of EARR need to be considered when analyzing them, the precipitation of EARR is more accurate than that of ERA5 for both two seasons. The EARR data presented here can be downloaded from https://doi.org/10.7910/DVN/7P8MZT for data on pressure levels and https://doi.org/10.7910/DVN/Q07VRC for precipitation.