A last question

Abstract

Diurnal variations of cloud cover and cloud vertical distribution are of great importance to earth-atmosphere system radiative budgets and climate change. However, thus far, these topics have received insufficient attention, especially on the Tibetan Plateau (TP). This study focuses on the diurnal variations of total cloud cover, cloud vertical distribution, and cirrus clouds and their relationship to meteorological factors over the TP based on active and passive satellite observations, reanalysis data, and CMIP6 outputs. Our results are consistent with previous studies but provide new insights. The results show that total cloud cover peaks in the afternoon, especially over the southeastern TP, but the spatial and temporal distributions of clouds from different datasets are inconsistent. To some extent, it could be attributed to subvisible clouds missed by passive satellites and models. Compared with satellite observations, the amplitudes of the diurnal variations of total cloud cover obtained by the reanalysis and CMIP6 models are obviously smaller. The CATS can capture varying pattern of the vertical distribution of clouds and corresponding height of peak cloud cover at middle and high atmosphere levels, although it underestimates the cloud cover of low-level clouds especially over the southern TP. Compared with CATS, ERA5 cannot capture the complete diurnal variations of vertical distribution of clouds and the MERRA-2 has a poorer performance. We further find that cirrus clouds, which are widespread over the TP, show significant diurnal cycle and spatial and temporal distribution characteristics, with peak cloud cover over 0.4 during 15:00–21:00 LT over the northeastern TP. Be different from tropic, where thin cirrus (0.03<optical depth<0.3) dominate, opaque cirrus clouds (0.3<optical depth<3) are the dominant cirrus clouds over the TP. The cloud cover of opaque cirrus reaches a daily maximum of ~0.24 over the northeastern TP at 15:00 LT, and are influenced by diurnal variations of the 2-m temperature and 250 hPa vertical velocity. Although subvisible clouds (optical depth<0.03), which have a potential impact on the radiation budget, are the fewest among cirrus clouds over the TP, the cloud cover can reach 0.1 during 21:00–03:00 LT, and their diurnal cycle is obviously consistent with that of the high-level relative humidity. Our results will help reduce uncertainties in simulations of diurnal variations of cloud cover in models and reanalysis data over the TP region.