The airborne two-dimensional stereo (2D-S) optical array probe has been operating for more than 10 yr, accumulating a large amount of cloud particle image data. However, due to the lack of reliable and unbiased classification tools, our ability to extract meaningful morphological information related to cloud microphysical processes is limited. To solve this issue, we propose a novel classification algorithm for 2D-S cloud particle images based on a convolutional neural network (CNN), named CNN-2DS. A 2D-S cloud particle shape dataset was established by using the 2D-S cloud particle images observed from 13 aircraft detection flights in 6 regions of China (Northeast, Northwest, North, East, Central, and South China). This dataset contains 33,300 cloud particle images with 8 types of cloud particle shape (linear, sphere, dendrite, aggregate, graupel, plate, donut, and irregular). The CNN-2DS model was trained and tested based on the established 2D-S dataset. Experimental results show that the CNN-2DS model can accurately identify cloud particles with an average classification accuracy of 97%. Compared with other common classification models [e.g., Vision Transformer (ViT) and Residual Neural Network (ResNet)], the CNN-2DS model is lightweight (few parameters) and fast in calculations, and has the highest classification accuracy. In a word, the proposed CNN-2DS model is effective and reliable for the classification of cloud particles detected by the 2D-S probe.

Vertical information about aerosols and clouds is vital to understanding aerosol transport, aerosol-cloud interactions, and pollution-weather-climate feedback so as to reduce uncertainties in estimating their climatic effects. The combination of sounding, lidar, aircraft, and satellite measurements is widely used to obtain the vertical information of aerosols and clouds. We used an aircraft measurement over southern Anhui, the upstream regions of Shanghai, on 1 November which conducted to ensure good air quality for the Third China International Import Expo to examine the vertical characteristics of aerosol and cloud microphysical properties and their variations before and after cloud seeding. Observations showed aerosols and clouds were vertically stratified. Most aerosols trapped within the boundary layer are small particles with sizes less than 0.12 µm. Aerosol number concentrations ( N a ) generally decreased with altitude in the cloudless atmosphere, with the largest particles occurring in 2500–3500 m due to dust transported from distant regions and high ambient humidity. Four separate cloud layers with unequal depths dominated by altostratus and nimbostratus appeared at different heights. The maximum cloud droplet concentration ( N c ) and the minimum cloud droplet diameter ( D c ) that appeared in the mid-level cloud (2246–2482 m) were 107.7 cm −3 and 4.03 μm, respectively, owing to the high proportion of hygroscopic particles. Hygroscopic particles played an important role in the growth of droplets and the activation of cloud condensation nuclei, especially under high ambient humidity. Cloud droplet size spectrum showed a unimodal distribution with a single peak at 5 µm in low- (970–1000 m) and mid-level clouds, but a trimodal distribution with peaks at 7 μm, 12 μm, and 17 μm in the mid-high- and high-level clouds, indicating the broadening of spectra with increasing altitude. An artificial seeding experiment was conducted in the high-level clouds. Big cloud droplets and ice crystals increased significantly after cloud seeding. Meanwhile, cloud particle populations showed less N c , larger D c , and a wider size spectrum. Our results suggest that the artificial precipitation experiment promoted rainfall to a certain extent and contributed to the removal of pollutants from upstream regions, which is beneficial to the air quality of Shanghai.

Identifying supercooled liquid water (SLW) in clouds is critical for weather modification, aviation safety, and atmospheric radiation calculations. Currently, aircraft identification in the SLW area mostly depends on empirical estimation of cloud particle number concentration ( N c ) in China, and scientific verification and quantitative identification criteria are urgently needed. In this study, the observations are from the Fast Cloud Droplets Probe, Rosemount ice detector (RICE), and Cloud Particle Imager (CPI) onboard a King Air aircraft during seven flights in 2018 and 2019 over central and eastern China. Based on this, the correlation among N c , the proportion of spherical particles ( P s ), and the probability of icing ( P i ) in supercooled stratiform and cumulus-stratus clouds is statistically analyzed. Subsequently, this study proposes a method to identify SLW areas using N c in combination with ambient temperature. The reliability of this method is evaluated through the true skill statistics (TSS) and threat score (TS) methods. Numerous airborne observations during the seven flights reveal a strong correlation among N c , P s , and P i at the temperature from 0 to -18°C. When N c is greater than a certain threshold of 5 cm -3 , there is always the SLW, i.e., P i and P s are high. Evaluation results demonstrate that the TSS and TS values for N c = 5 cm -3 are higher than those for N c < 5 cm -3 , and a larger N c threshold (> 5 cm -3 ) corresponds to a higher SLW identification hit rate and a higher SLW content. Therefore, N c = 5 cm -3 can be used as the minimum criterion for identifying the SLW in clouds at temperature lower than 0°C. The SLW identification method proposed in this study is especially helpful in common situations where aircraft are equipped with only N c probes and without the CPI and RICE.

西北区域人工影响天气能力建设项目设计建设的新舟60增雨飞机是集催化作业、云宏微观探测以及实时通信与综合集成显示功能于一体的国家高性能人工增雨飞机.该飞机系统建设中充分汲取了前期东北区域国家增雨飞机研制成果与经验,并在系统供电管理、催化作业能力、探测系统集成、卫星通信功能、设备系统和操作台布局等方面进行了针对性设计改进,通过便捷操作实现对任务系统供电与各分系统的集中控制、综合显示、数据存储共享,并且机载探测、催化、通信各分系统均采用双重或多重备份设计,既保证了机载任务系统先进性设计,同时保证任务系统运行稳定可靠和各项功能的完整实现.

Currently, the most direct and effective way to obtain cloud precipitation microphysical characteristics is from in-situ measurements acquired by airborne imaging probes. There are many studies based on CIP (cloud imaging probe) and PIP (precipitation imaging probe) detection data, which are mostly based on the limited output from the software PADS (Particle Analysis and Display System) provided by DMT (Droplet Measurement Technologies). Since PADS only outputs the second-by-second statistical results rather than the detailed particle-by-particle information, it greatly limits the deep mining and analyzing of cloud particle image data. Besides, particle shapes in previous studies are mainly classified through naked-eye observations, which is time consuming, subjective, and unreliable to conduct statistical analysis on thousands of cloud particle images. Therefore, it is impossible to calculate the hydrometeor content based on mass-dimension relationships for particles of different shapes in ice or mixed cloud observations.The operation principle of airborne two-dimensional optical array probes is introduced. Then techniques of recognition and elimination of shattering particles and fake particles are illustrated in detail. Particle shapes are divided into 8 types (tiny, linear, aggregated, graupel, spherical, plate, dendritic and irregular) based on geometric characteristics of particle shapes. Statistical characteristics of different cloud particle shapes and their areas are analyzed by using gray CIP data detected in three wintertime stratiform clouds in Henan Province. The recognition of particle shapes are basically consistent with results through naked-eye observations, and also consistent with dominant particle shapes in each temperature range obtained by previous studies. The hydrometeor content obtained using the mass-dimension relationship for particles of different shapes is compared with that from treating all particles as spherical liquid particles (i.e., the algorithm used by PADS). It is found that when all particles are treated as spherical liquid particles, the hydrometeor content is roughly one magnitude higher than that from considering different particle shapes, indicating the technique of particle shape classification can improve the accuracy of hydrometeor content in ice or mixed clouds. In addition, some matters needing attention in the use of two-dimensional particle image data are pointed out to ensure proper use of two-dimensional particle image data.

为了了解机载探测不同数据源之间差异,检验飞机探测数据质量,结合新舟60增雨飞机2018年10月21日一次飞行案例,对该飞机平台和飞机任务系统不同机载设备对关键飞行参数与气象要素的观测对比分析.结果 表明:飞机平台全球卫星定位系统(GPS)和任务系统北斗卫星导航系统(BDS)、机载综合气象测量系统(AIMMS-20)三套定位源的经度、纬度和海拔高度定位偏差较小,飞机气压高度表观测海拔高度则明显低于任务系统BDS和AIMMS-20.飞机平台大气数据系统(ADS)与任务系统AIMMS-20、云粒子图像探头(CIP)观测的真空速、环境气压、温度以及相对湿度各参数变化趋势一致.ADS与AIMMS-20真空速观测值非常接近,AIMMS-20对环境风速瞬时变化响应更灵敏,CIP探测真空速明显小于ADS和AIMMS-20,平均偏慢约10 m·s-1;AIMMS-20和CIP环境气压变化趋势完全一致,且观测值非常接近,ADS环境气温比AIMMS-20平均偏低1.4℃,偏低最大时观测存在明显逆温.CIP环境气温比AIMMS-20平均偏低0.6℃,环境湿度比AIMMS-20平均偏低8.6％.机上不同设备对环境与气象参数的观测差异,一方面因机载设备安装位置不同所致,另一方面也受大气与云结构不均匀的影响.飞机平台与任务系统不同设备观测对比分析,不仅为云物理机载探测数据合理应用提供指导,还能够为国家和地方人工影响天气飞机机载设备系统集成设计提供重要技术支持.

考虑到下行负极性地闪过程中大多上行正先导发展时不分叉的观测事实,基于已有的闪电先导二维随机模式,改变上行正先导的模拟方案,使其发展时不产生分叉,并对雷击高建筑物过程中下行先导与上行连接先导(up-ward connecting leader,UCL)之间头部-头部连接和头部-侧面连接(侧击)的两种形态进行模拟.以高度为440 m的建筑物为例,通过改变下行先导始发点(高度为1000 m)与高建筑物的水平距离,模拟高建筑物上雷击过程中先导之间的连接过程,结果表明:当下行先导始发点与高建筑物的水平距离从0增加到700 m时,UCL的长度呈持续增大趋势,UCL受侧击的概率总体上呈先增大后减小的趋势(在距离为500m时达到最大值58％),侧击时UCL连接点以上的部分占整个UCL长度的比例总体呈持续增大趋势(13％～49％).