Regionalization of the Summertime Planetary Boundary Layer Height in Comparison with Various Reanalysis Datasets Over China

Abstract

The knowledge remains poor concerning the regionalization of planetary boundary layer height (PBLH) in a huge domain as large as China. Six regions were classified in this study, determined by K-means clustering of the radiosonde-derived PBLHs in summer over China and related influential factors. Besides, a comprehensive intercomparison analysis is performed against serveral reanalysis products, including ERA5, FNL and CERA. The PBLHs exhibit large spatiotemporal dependence. During a day, the lowest PBLHs are observed at 0800 BJT, followed by those at 2000 BJT. In addition, the PBLHs in China reach the peak at 1400 BJT, showing a "northwest high-southeast low" spatial pattern, with the largest PBLHs over Inner Mongolia (an average reaching 2259 m), probably due to the dry soil and high altitudes. Nationwide, ERA5 has the smallest biases and the closest relation to the sounding observations among the three reanalysis, suggesting that it is robust in estimating PBLHs. The boundary layer at 1400 BJT is further classified into convective (CBL), neutral (NBL) and stable boundary layer (SBL) according to the thermodynamic conditions. ERA5 is closely related to the sounding observations for CBL and NBL, bearing a correlation exceeding 0.7 over most regions. However, all the reanalysis products severely overestimate SBL heights. Comparisons further indicates that the PBLH biases of reanalysis products are partly due to misrepresentation of PBL thermal types, which is probably attributed to the less satisfactory simulation of cloud effects in models. To our best knowledge, this study presents the first regionalization of high-resolution radiosonde-derived PBLHs and provides a quantitative assessment of the three mainstream reanalysis products in China.

Keywords: PBLH regionalization, K-means clustering, high-resolution radiosonde, mainstream reanalysis products, thermodynamic conditions