Old descriptive diaries are important sources of daily weather conditions before modern instrumental measurements became available. A previous study demonstrated the potential of reconstructing historical weather at high temporal resolution by assimilating cloud cover converted from descriptive diaries. However, cloud cover often exhibits a non-Gaussian distribution, which violates a basic assumption of most data assimilation schemes. In this study, we applied a Gaussian transformation (GT) approach for cloud cover data assimilation and conducted observing system simulation experiments (OSSEs) using 15 randomly selected observation points over Japan. We performed experiments to assimilate cloud cover with large observational errors using the Global Spectral Model (GSM) and a local ensemble transform Kalman filter (LETKF). Without GT, temperature and zonal and meridional wind exhibited deterioration compared to the experiment assimilating no observations. By contrast, the 2-month root mean square error (RMSE) of zonal wind, meridional wind, temperature, and specific humidity at mid-troposphere were improved by 8.7%, 5.1%, 4.2%, and 1.4%, respectively, through GT. Among two-dimensional variables, the 2-month RMSE of total cloud cover, surface pressure, rainfall, and downward solar radiation were improved by 2.2%, 5.2%, 27.6%, and 4.3%, respectively. We further demonstrated that the effect of GT was more pronounced on clear days. Our results show the potential of GT in high-resolution historical weather reconstruction using old descriptive diaries.