Arranging 3 years’ worth of airborne precipitation particle images to construct a precipitation particle image dataset in Shan⁃ dong Province. Building a precipitation particle recognition model based on EfficientNet convolutional neural network, named PREN (Precipitation particle Recognition model based on EfficientNet convolutional neural Network).The accuracy rate is 98%, and the multimodel and multi-index evaluation and comparison experiments verify that PREN demonstrates excellent robustness and generalization ability. Taking typical stratiform-cumulus mixed cloud precipitation as two examples (total 3 time periods), PREN is applied to the par⁃ ticle characteristics analysis of generating cells. Combined with airborne Ka-band cloud radar and DMT particle measurement system, an analysis conducted on the shape proportion of precipitation particles inside and outside the generating cells and indifferent intensity generating cells, revealing the precipitation mechanism. The results show that the shapes of precipitation particles in the generating cells are mainly spherical, needle-like, irregular and columnar. Precipitation particles outside the generating cells are mostly spherical and needle-like. The cloud microphysical parameters in the generating cells with different intensities vary. The proportion of graupel and needle particles in the precipitation maturity stage is higher than that in the dissipation stage. The average chord length of precipi⁃ tation particles in the maturity stage is 415 µm. While the average chord length of particles in dissipation stage is 367 µm. The par⁃ ticles on the top of generating cells are mainly spherical and hexagonal, primarily growing through the process of deposition. The ratio of irregular particles and columnar particles in the 0 ℃ are increasing, and the melting process and dynamic conditions favor aggregation and growth, forming irregular particles, while columns mainly originate from the upper levels of the atmosphere.

The research on the distribution of cloud water content and its evolution rules has important significance for the exploitation and utilization of regional cloud water resources. The paper analyzed the temporal variation characteristics of liquid water path (LWP) and integrated water vapor (IWV) in central Guanzhong Plain by using observation data of MWP967KV ground-based microwave radiometer at Jinghe station of Shaanxi Province from October 2017 to December 2020. Combined with ground precipitation and Doppler weather radar observation data, the development and evolution characteristics of water vapor and liquid water before precipitation in various cloud systems were compared by some cases study. The results indicate that the IWV exhibits obviously seasonal variations in central Guanzhong Plain, with the highest in summer, followed by autumn and spring, and the lowest in winter. Specifically, the peak appears in July, and the valley appears in December. The LWP is higher in autumn and summer, in winter it is the lowest. Notably, the peak is in September, and the valley is in December. The distribution of the IWV and LWP exhibits a single peak and single valley pattern over the course of a day, but the occurring time of their peak and valley is different. The diurnal maximum of the IWV occurs from 07:00 to 08:00 in summer and autumn, 23:00 in spring and 13:00 in winter, while the diurnal minimum of the IWV occurs at about 12:00 in spring, summer and autumn, 22:00 in winter. The diurnal maximum of the LWP occurs from 07:00 to 09:00 in spring, summer and autumn, while in winter it is slightly late (10:00). The diurnal minimum of the LWP appears at the nighttime in all seasons. The growth time of cloud water content before precipitation is different for different types of cloud systems. On average, the development time of stratiform cloud systems is 15.6 hours, and for other cumulus cloud systems it is 9.0 hours. In the initial stage, the IWV in both cloud systems varies prior to the LWP, and the fluctuation amplitude is increasingly violent as precipitation approaches. Additionally, the LWP in both cloud systems firstly exhibits a sudden violent increase before the rainfall being triggered, and the IWV and LWP in stratiform cloud system vary greatly in different seasons as precipitation is triggered. In the afternoon, the duration of strong convection developing is short, with an average time of 30 minutes. In the initial stage of development and before precipitation, the LWP varies and jumps sharply at the first.

The summer monsoon transition zone in China is one of the regions with strong land-atmosphere interaction in the world, and it is also an area where extreme weather disasters are frequent and easy to cause serious economic losses. Further understanding of land-atmosphere interaction in the transition area will help to improve the disaster prevention and mitigation ability of this region. Based on the research results of the summer monsoon transition area related projects carried out by the Key Laboratory of Drought Climate Change and Disaster Reduction of China Meteorological Administration in recent years, this paper systematically summarizes the new progresses of land-atmosphere interaction in the summer monsoon transition zone, including the spatio-temporal distribution law of land-atmosphere interaction in the transition region, the new characteristics of the response of land surface water budget to summer monsoon, the spatio-temporal variation characteristics and development mechanism of the boundary layer, the influence of monsoon and land-atmosphere interaction on regional climate in the transition zone, the new progress of land-atmosphere interaction on crop yield in the transition zone and new schemes for parameterization of multi-factor and multi-scale kinetic roughness. According to the development trend of land-atmosphere interaction research in the summer monsoon transition zone, it is proposed that the multi-scale dynamic response of land-atmosphere interaction to summer monsoon should be explored in the future, and the climatic dynamic relationship between surface processes and key physical quantities in the atmospheric boundary layer should be established on the basis of the research on the response rule of land-atmosphere exchange multi-cycle process to the annual cycle of summer monsoon in order to improve and enhance the simulation of regional climate models in the future. This work is of great significance to promote the research of land-atmosphere coupling process in China, which can provide scientific and technological support for disaster prevention and mitigation in the summer monsoon transition zone in China.

Based on the data from three wind profile radars, the rain and snow weather process in the central part of Shandong Province from February 12 to 13, 2016 was analyzed. The results are as follows: (1) This rain and snow weather was the return-flow precipitation mainly caused by the combination of low-altitude vortex and surface cyclone. Heavy precipitation occurred in the cold and north side of the cyclone center. The temperature on 925 hPa and near-surface  dropping below 0 ℃ was the important indicator of the conversion from rainfall to snowfall. (2) The change of wind field below 1 km was the signal of precipitation start. The change of strength of low-layer cold pad determined the transition of precipitation phase. The maximum detection height of the radar during snowfall was significantly lower than that during rainfall. The low-altitude wind-shear index would increase significantly at the beginning and end of precipitation. The small increase of the low-altitude wind-shear index during the continuous rainfall period corresponded to the increase of rainfall intensity. (3) The vertical velocity of the wind profile radar could not only reflect the change of precipitation intensity, but also reflect the change of precipitation phase. The vertical velocity near ground gradually decreased during phase transition. (4) Both structure constant of atmospheric reflective index and signal-to-noise ratio showed a decreasing trend with height, and the changes of them could reflect the change of precipitation intensity. They all had zero-level bright band characteristics, and the disappearance of bright band corresponded to the beginning of snowfall.

The variation tendency of soil water content and the relationship between soil drought and meteorological drought were analyzed
based on the daily and the real － time observed data at Nanjing from July 2010 to June 2011，and the meteorological drought composite
index and the soil relative humidity index were calculated by using meteorological observations at Nanjing and daily precipitation，
10 － 100 cm soil moisture content at the experimental station constructed by Nanjing University of Information Science ＆ Technology
and Institute of Arid Meteorology，CMA． The results show that Nanjing suffered a severe dry weather in winter of 2010 and spring of
2011，on November 5，2010，mild drought began to appear and changed to moderate drought on 12 Nov and severe drought on 28
Nov． ，and then maintained at moderate and severe drought． Soil relative humidity reached moderate drought on 13 November，after 15
days of continuous severe meteorological drought，the soil moisture reached severe drought on 2 May， 2011． The trends of meteorological
and soil drought were basically consistent，but the extent of meteorological drought was more serious than that of soil drought，and the
starting and ending time of soil drought lagged 1 － 3 days，and the drought development lagged behind more than 5 days． Weather and
surface soil had higher sensitivity to precipitation，while the middle soil drought lasted long． In addition，when the meteorological
drought duration reached 50 － 60 days，water would add from the deep layer to the top layer of soil．