The rapid warming and consequent retreat of glaciers across the Tibetan Plateau (TP) have given rise to the debate on the ability of the atmospheric water supply to alleviate the depletion of surface water storage. We investigate long‐term changes in atmospheric water vapor balance across the TP using 40‐year fifth generation European Centre for Medium‐Range Weather Forecasts (ECMWF) (ERA5) reanalysis. Precipitation, water vapor convergence, and evaporation generally maintain an equilibrium but with different long‐term variation trends: 0.68, −0.18, and 0.69 mm/a, respectively. Results suggest the inability of the water vapor arriving from outside the TP to effectively replenish the surface water storage. Despite of huge changes in atmospheric water vapor balance during summer, the risk of water storage depletion is brought by other three seasons, especially the autumn. The climatology and long‐term trends of water vapor balance exhibit strong variation across the regions of TP. The surrounding areas of Yarlung Zangbo Grand Canyon experience sharp decrease in water vapor convergence, thus reducing precipitation. Moreover, increasing evaporation is expected to severely loss of surface water storage. For the Three Rivers Source Region with no significant changes in total precipitations, decrease in water transported from outside of TP overlaps increase in evaporation results in the possible depletion of surface water storage. Brahmaputra basin, inner TP, and Qilian Mountain exhibit significant wetting trends due to increases in both convergence of water vapor flux and evaporation. Above regional and seasonal characteristics of water vapor, balances across the TP are attributed by inhomogeneous variation of atmospheric heat source and complex changes of atmospheric circulations.

Tibetan Plateau (TP), as the "third pole" of world, has experienced significant and rapid warming over the past several decades with a warming rate of about twice the global rate (Chen et al., 2015), even during the period of the global hiatus (Duan & Xiao, 2015;You et al., 2016). Such rapid warming has caused glacier retreat and permafrost degradation (Bibi et al., 2018;Yao et al., 2019) and has been proved to be partly related to the changes of cloud properties (e.g., reduced total and low cloud covers during daytime allowed more solar radiation to reach the surface) (

Cloud droplets homogeneously freeze at temperatures below about −40°C without ice nucleating particles (INPs), but atmospheric aerosols can act as INPs so that supercooled droplets may freeze somewhat between −40°C and 0°C. This process is called heterogeneous freezing (Pruppacher & Klett, 1978). Thus, clouds can be comprised of liquid droplets, ice crystals or a mixture of the two (i.e., mixed-phase) for this temperature range. Such thermodynamic phase composition (e.g., supercooled water cloud fraction, SCF) plays an important role in the cloud precipitation efficiency and lifetime due to different optical properties and cloud feedback effects between liquid and ice particles (Choi et al., 2014;Jiang et al., 2000). Therefore, the proper partitioning of cloud phase (CP) can significantly affect the radiation budget and climate sensitivity (

Clouds are considered as an important factor of regulating weather and climate systems due to their key role in controlling the energy budget, hydrological and general circulation (Stephens, 2005;Xu et al., 2022;Zhang & Jing, 2016). Multilayer cloud systems frequently appear in the atmosphere (Li et al., 2011(Li et al., , 2015 and can significantly alter the radiative heating rate profile. Their radiative effect is complex and strongly depends on the vertical distribution of cloud layers, cloud heights, cloud optical depths (CODs) and cloud phases ( Johansson et al., 2019;Li et al., 2011). As an important microphysical process, natural seeder-feeder mechanisms in multilayer clouds can also influence precipitation and promote extreme precipitation, but this process is difficult to be reproduced in models (Avramov & Harrington, 2010). In the context of global warming, therefore, more comprehensive detection and research into multilayer clouds are necessary to understand the global and regional energy budget and water cycle.Previous studies have shown that overlying clouds can affect the underlying clouds via their longwave radiative effect (Adebiyi et al., 2020;Christensen et al., 2013;Proske et al., 2021). On average, the overlying clouds can decrease the cloud-top radiative cooling rate of the underlying stratocumulus (Sc, usually occurs below 2 km and is defined as a low-level cloud system whose dynamics are primarily driven by convective instability caused by cloud-top radiative cooling), thus reduce turbulent mixing, and inhibit the vertical development and rain rate of the Sc (

Precipitation change is determined by changes in frequency and intensity. Based on hourly and daily precipitation data sets in China (1961–2014) and CMIP6 model, this study found that observed inconsistency in the frequency trend at the two time resolutions (hourly and daily) reflected precipitation becoming more concentrated at 13.4% of stations, resulting in the specific precipitation phenomenon of “more hours in one day, but fewer days.” However, models exhibited opposite precipitation phenomenon at 16.5% of stations and cannot reproduce the widespread increase in the proportion of extreme precipitation amounts (PA) in the total PA, especially at an hourly resolution. Although stations with significant trend from CMIP6 approximately twice the observation, both CMIP6 and observations showed that frequency and intensity changes dominated the total PA trend at hourly and daily time resolutions, respectively. These results demonstrated that daily precipitation observations could not fully capture the important features of short‐duration rainfall.