Microphysical characteristics of precipitating cumulus cloud based on airborne Ka-band cloud radar and droplet measurements

Li Wi , , , Mnyu Hun , , Ron Zhn , Yuhun Lü , Tunji Hou , Hnhi Li ,Dlon Zho , , , , Wi Zhou , , Yun Fu

a Beijing Weather Modification Center, Beijing, China

b Key Laboratory for Cloud Physics of China Meteorological Administration, Beijing, China

c Wuqing Meteorological Observatory of Tianjin, Tianjin, China

d Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China

e Zhejiang Meteorological Observatory, Hangzhou, China

f Beijing Key Laboratory of Cloud, Precipitation and Atmospheric Water Resources, Beijing, China

g Beijing Meteorological Bureau, Beijing, China

ABSTRACT Based on cloud-probe data and airborne Ka-band cloud radar data collected in Baoding on 5 August 2018, the microphysical structural characteristics of cumulus (Cu) cloud at the precipitation stage were investigated. The cloud droplets in the Cu cloud were found to be significantly larger than those in stratiform (STF) cloud. In the Cu cloud, most cloud particles were between 7 and 10 μm in diameter, while in the STF cloud the majority of cloud particles grew no larger than 2 μm. The sensitivity of cloud properties to aerosols varied with height. The cloud droplet effective radius showed a negative relationship with the aerosol number concentration (Na) in the cloud planetary boundary layer (PBL) and upper layer above the PBL. However, the cloud droplet concentration (Nc)varied little with decreased Na in the high liquid water content region above 1500 m. High Na values of between 300 and 1853 cm ? 3 were found in the PBL, and the maximum Na was sampled near the surface in August in the Hebei region, which was lower than that in autumn and winter. High radar reflectivity corresponded to large FCDP (fast cloud droplet probe) particle concentrations and small aerosol particle concentrations, and vice versa for low radar reflectivity. Strong updrafts in the Cu cloud increased the peak radius and Nc, and broadened cloud droplet spectrum; lower air temperature was favorable for particle condensational growth and produced larger droplets.

1. Introduction

The warm rain formation process of cumulus (Cu) convection refers to the formation process of raindrops without ice particles. A large number of observations and studies have found that warm Cu clouds developing to a certain height over the ocean will lead to the formation of warm rain and precipitation. Over the tropical oceans, radar reflectivity data have shown that precipitation occurs after 15—20 min of development of shallow convective cloud ( Szumowski et al., 1997 ; Minor et al.,2011 ). Over land, it is difficult for raindrop embryos and warm raindrops to form via random collision because of the high concentration of aerosols and cloud droplets and the narrow cloud droplet spectrum( Rogers and Yau, 1989 ). Many previous studies have shown that aerosols can change the microphysical properties of clouds ( Twomey, 1977 ;Garrett et al., 2004 ; Wang et al., 2014 ; Qiu et al., 2017 ). The condensational growth of cloud droplets is an important factor affecting the formation of raindrop embryos. That is, all factors restricting supersaturation are important factors affecting raindrop embryo formation.

To date, there is no universally accepted answer to the mechanisms of the formation of warm raindrops ( Knight et al., 2002 ). To reveal the formation mechanisms of raindrop embryo particles, cloud physicists have proposed many theories. Among them, one theory is that large cloud droplets are formed after meganucleus activation( Woodcock et al., 1971 ; Jensen and Lee, 2008 ). Another theory is that turbulence is helpful to improve the coalescence efficiency of cloud droplets ( Woods et al., 1972 ; Shaw, 2003 ), and the entrainment and nonuniform mixing of Cu side boundaries are helpful in forming larger cloud droplets ( Baker et al., 1980 ), thus forming raindrop embryos.

Joint observations from aircraft, radar, and numerical simulations are the main ways to study the formation of warm rain. From the results of the Small Cumulus Microphysics Study ( Lasher-Trapp et al., 2001 )and the Rain in Cumulus over the Ocean experiment ( Rauber et al.,2007a , b ), the formation of warm rain was found to include two processes: the formation of raindrop embryos and the collision-based growth of raindrop embryos and cloud droplets. Therefore, to study the formation processes of warm rain, on the one hand it is necessary to study the process of raindrop embryo formation —especially the microphysical and dynamic processes affecting supersaturation, as well as the environmental conditions and time required for embryo formation;whilst on the other hand the rapid outbreak of warm rain after embryo formation has been studied via early X-band radar observations, and,by comparing the raindrop echo of meganucleus formation simulated by the model with the observations of X-band radar, it was found that meganuclei play an important role in the formation of raindrop embryos and raindrops ( Lasher-Trapp et al., 2001 ).

Aircraft observations can help elucidate the relationship between aerosols near cloud bases and cloud microphysical properties. Although,a problem in this respect is that during aircraft cloud microphysical observations it is difficult for ground-based radar data to match with those of cloud-physics probes in time and space, resulting in large errors. However, the first airborne Ka-band cloud radar (KPR) in China is able to observe the macrostructure of cloud particle characteristics in precipitation clouds during the process of aircraft penetrating clouds,and the data can then be combined with those of other microphysical instruments to obtain microphysical observational datasets. At present,there have been few studies on the microphysical characteristics of KPRdetected clouds in China ( Pazmany and Haimov, 2018 ; Zhang et al.,2020 ). KPR is a new type of meteorological remote sensing equipment that has high frequency and high spatial resolution. It can observe the macrostructure of cloud particles in precipitating clouds during the process of aircraft passing through the clouds.

In this study, we investigated the microstructures of precipitating clouds in Baoding in summer. With the use of coupled detection data of the Cu cloud system on 5 August 2018, combined with radar data, the characteristics of cloud particles measured during aircraft detection and the corresponding condensation mechanisms and microphysical processes/properties were analyzed in detail to elucidate the structure of the Cu cloud system in the study area.

2. Data description

The King Air research aircraft owned by the Beijing Weather Modification Office (Beijing WMO), equipped with both in-situ and remotesensing probes, including a fast cloud droplet probe (FCDP), twodimensional stereo optical array imaging probe (2D-S), passive cavity aerosol spectrometer probe (PCASP), and KPR, was employed to conduct aircraft observations. The exploration flight experiment included vertical and horizontal measurements at different levels in the target area. The 2D-S probe consisted of two independent, identical cloud particle probes in one housing, called the ‘H’ (horizontal) and ‘V’ (vertical)channels. The FCDP records particles with a size range from 2 to 50μm and the 2D-S images particles from 10 to 6400μm. The PCASP measures aerosol particles in 30 size bins from 0.1 to 3.0μm.

The KPR installed on the King Air research aircraft was originally developed by the University of Wyoming ( Pazmany and Haimov, 2018 ).Specifications for the Beijing WMO KPR are summarized in Table 1 .KPR includes two antennas, one pointing up and one pointing down,providing simultaneous measurements above and below the aircraft.

Table 1 KPR specifications.

Table 2 Horizontal flight leg details, including the aircraft, beginning and ending times, altitude, and cloud temperature.

3. Flight experimental design

The King Air research aircraft departed from Shahe Airport to observe the convective cloud precipitation system, based upon which this study investigates the variation in the cloud particle size distribution(PSD) with the aerosol number concentration (Na) in shallow Cu cloud with sufficient liquid water content (LWC). A rigorous experimental design was established to analyze the precipitation processes of the convective cloud.

The observation area of this experiment was mainly over Baoding.On 5 August 2018, the aircraft carried out one detection campaign of the cloud system, including seven horizontal flight legs ( Fig. 1 ). Generally,the method of downward observation from the cloud top was adopted.However, for small Cu clouds, the disturbance of the aircraft itself may have an impact on microphysical processes, and so the downward observation to the cloud base scheme was used. All flight legs targeting the same cloud case were executed consistently, except that they occurred at different horizontal levels. The aircraft carried the cloud microphysical and KPR instruments for simultaneous horizontal detection of the cloud system over Baoding. For this study, six horizontal flight legs above 500 m and below 2500 m (above sea level) during the aircraft descent were selected and used mainly to study the PSD near the boundary layer. The corresponding horizontal flight information is listed in Table 2 .

4. In-situ aircraft measurements

4.1. Cloud droplet sizes and concentrations

Fig. 1 (b) shows the variation in flight altitude and distribution of cloud droplets (based on the FCDP data). After the aircraft departed from Shahe airport between 0903 and 1003 CST (China Standard Time),most particle number concentrations were 0 cm?3, which indicated the aircraft was outside the cloud; only between 0907 and 0912 CST was the particle number concentration approximately 3000—4200 cm?3. In this period, the instrumentation did not detect clouds, and only after entering Baoding was there a small Cu cloud. The flight altitude of the aircraft was below 2200 m, and several vertical horizontal detection measurements were performed from high to low levels (see Table 2 ).The maximum particle number concentration did not exceed 7000 cm?3.The concentration of cloud droplets detected in the Cu cloud was much higher than that detected in stratiform (STF) cloud. The number concentration of cloud droplets is less than 102cm?3in STF cloud ( Wei et al.,2021 ).

Fig. 1. (a) Three-dimensional trajectory diagram of the aircraft. (b) Aircraft flight height and particle number concentration change with time (based on the FCDP data).

Fig. 2. Based on the FCDP data of the King Air research aircraft, the variation in (a) cloud droplet concentration (Nc), (b) average diameter (Da), and (c) liquid water content (LWC).

The atmospheric temperature was above 15°C (see Fig. 2 ) for cloudtop heights below 3000 m. All these clouds were in pure liquid phase without the presence of ice particles, and the temperature at all cloud heights exceeded 0°C. Profiles of both aerosol and cloud properties were obtained from the aircraft observations. Fig. 2 shows the vertical distribution of cloud droplets measured by the King Air research aircraft (a 5-s average processing of the FCDP data).

The cloud droplet concentration (Nc) detected by the FCDP with a broader size range was as large as 103cm?3, mainly in the Cu cloud.The FCDP detected the concentration of aerosols and cloud particles below the 16°C layer (see Fig. 2 ). The clouds with the maximum particle concentration ranged from 2000 to 3000 cm?3at temperatures of 16°C—24°C (as shown in Fig. 2 (a)). There was a dry layer between 24°C and 25.5°C, most of which had an LWC value of 0 g m?3, indicating the aircraft was outside the cloud. There was a high region of LWC between 16°C and 24°C, and the extreme-value region of LWC was close to 1.23 g m?3. There were large vertical variations of cloud droplet average diameter (Da), which were around 5—11.5μm for the low cloud layer below 1200 m and 1—13.5μm above the planetary boundary layer(PBL) height. Decreased temperature with height helped increase the Da through the cloud condensation process.

Fig. 3. Particle spectrum detected by the FCDP.

Fig. 4. Variation distribution of (a) aerosol number concentration (Na) with sizes between 0.1 and 3 μm measured by the PCASP and (b) cloud droplet effective radius (Re).

Variations existed for both Nc and LWC, with values from a few to less than 3000 cm?3, and from 10?7to 1.23 g m?3, respectively.Yang et al. (2019) found that if the LWC is high and the aerosol amount not too large, both the Nc and the cloud droplet effective radius (Re)tend to increase with increasing aerosol concentration; whereas, if the LWC is low, or high but with an excessive aerosol amount, the Nc increases but the Re decreases with increasing aerosol concentration. The LWC in the PBL was bimodal below 1200 m (22°C), whilst above the PBL it generally varied little with height between 1500 and 3000 m, which is inconsistent with in-situ observations for liquid-phase clouds in relatively clear regions ( Zhao et al., 2019 ). In our study, above 1500 m,the Nc and LWC varied little, the cloud Da and the aerosol Re increased slightly, while the aerosol Na decreased with increasing height. The reduction in aerosol had little effect on the LWC, which mainly depended on the Nc and cloud Da.

The cloud droplet spectra (refer to Fig. 3 ) measured with the FCDP were of the monotonically decreasing type. The highest Nc was observed for the 7-μm cloud droplets at all flight altitudes. The concentration of particles less than 5μm was 0 cm?3μm?1. Regarding particles larger than 5μm, the concentration (plotted on logarithmic axes) decreased linearly with size (plotted on linear axes), suggesting that cloud droplets followed an exponential PSD. The particle concentration detected by the aircraft during detection increased with decreasing altitude over 1.52 km high and then decreased with decreasing altitude to 0.56 km. The concentration of small cloud droplets reached a maximum value during legs 3 and 4, which was approximately 100 cm?3μm?1. Compared with other legs, during leg 3, the particle concentration of cloud droplets larger than 5μm increased, but the concentration during leg 6 was much lower than that during leg 3. The cloud Da mostly varied between 5μm and 15μm, and the concentration of cloud droplets larger than 20μm reached 0.1 cm?3μm?1. The Nc was less than 100 cm?3μm?1in the Cu cloud, while that in STF cloud is usually larger than 100 cm?3μm?1( Wei et al., 2021 ).

Fig. 5. (a) Variation distribution of cloud number concentration with temperature, and (b) diameter and temperature variations with time (detected by the 2D-S probe).

4.2. Aerosol properties

For heights below the PBL height of approximately 1200 m (temperature was higher than 22°C), the Na was high, with values between 300 and 1853 cm?3, even for sizes ranging between 0.1 and 3μm (a 5-s average processing of the PCASP data). The Na with a broader size range in the bins can be as large as 104cm?3in North China. The amount of Na detected below the PBL increased with increasing temperature.Different from the Na, there was a large vertical variation in the cloud droplet Re ( Fig. 4 (b)). The cloud droplet radii were approximately 0.26—0.36μm for the low cloud layer with tops below 1200 m, and they were approximately 0.23—0.46μm above the PBL height (temperature was lower than 16°C). This could have been a result of vertical changes in both aerosols and temperature. Droplet radii increased through cloud condensation processes, in which decreasing temperature with height increased droplet radii. The decrease in aerosols made the clouds thinner and the droplets larger in size, but they did not compete too much with each other for water.

Fig. 6. Reflectivity (dBZ) from the flight and spherical particle images recorded with the 2DS ‘H’ channel on 5 August 2018 at (a) 1042—1043 CST and (b) 1049—1050 CST.

The heavy Na could significantly change the cloud properties within the atmospheric PBL. The aerosol concentration with a broader size range, such as between 0.01 and 10μm, can be as large as several 104cm?3at the surface under clear-sky conditions in North China, and can be even larger in winter and spring over the Beijing—Tianjin—Hebei region ( Hu et al., 2005 ; Wang et al., 2012 ; Zhang et al., 2014 ). Although the aerosol concentration in summer is not as large as that in February and September ( Liu et al., 2009 ; Zhao et al., 2018 ), a high-level aerosol number density of 1853 cm?3was detected by PCASP with the same size range of 0.1—3μm near the surface from August in the Hebei region.

4.3. KPR data and 2D-S particle features

The measurements illustrated in Fig. 5 (a) were collected using the 2D-S probe in the Cu cloud investigated from 1004—1058 CST during this project. This KPR example and diameter detected by the 2D-S probe( Fig. 5 (b)) was chosen because the particle image of the 2D-S probe was abnormal (no image perhaps because the aircraft was outside the cloud); only the images from 1042 to 1058 CST could be seen. Lower in the cloud (temperature was about 29°C), where there were higher concentrations of large cloud particles, the distribution peak was 2646 L?1, caused by the fragmentation of raindrop particles in the upper layer,and the main particle concentrations in other temperature layers were between 400 and 660 L?1. Between 1042 and 1048 CST, the particle maximum diameter measured by 2D-S was no more than 400μm, and most values were less than 100μm. In leg 6, the particle maximum diameter was larger than in leg 5, and the maximum particle diameter reached 800μm. All these clouds were in pure liquid phase without the presence of ice particles, and almost all had spherical particles (see Fig. 6 ).

The KPR dataset was collected on 5 August 2018 at 1042—1043 CST and 1049—1050 CST in Baoding, approximately 150 km southwest of Beijing. According to the radar echo, the aircraft was flying through convective cloud approximately 3 km above sea level. The radar pulse repetition frequency was constant at 20 kHz, transmitting alternating pairs of pulses to the up and down antennas. The results from the flight are shown in Fig. 6 . The flight altitude of the aircraft was approximately 0.88 km from 1042 to 1043 CST, the average particle number concentration detected by the FCDP was 16.73 cm?3, and the average Na concentration detected by the PCASP was 1411.47 cm?3. The average particle number concentration detected by the FCDP was 1.66 cm?3, the average Na concentration detected by the PCASP was 1561.45 cm?3at approximately 1050 CST, and the aircraft flight altitude was approximately 0.56 km. The intensity of cloud development in the two periods was similar; the maximum reflectivity in the Cu cloud detected by KPR was less than 25 dBZ, and the reflectivity of the radar echo near the ground could reach 35 dBZ, which was larger than that inside the Cu cloud. The ground echo was false, caused by ground reflections.

5. Conclusions

Based on the cloud microphysical and KPR probe observations, small particles (diameter from 5 to 10μm) accounted for the majority of all particles in the Cu cloud at each level. The slope of the spectra of the particles with a diameter larger than 7μm decreased monotonically.Cloud particles less than 2μm dominate in STF cloud ( Hou et al., 2010 ;Wei et al., 2021 ), while cloud particles with diameters mainly between 7 and 10μm were sampled in the Cu cloud in this study, which is significantly larger than those found in STF cloud.

The LWC in the Cu cloud was larger than that found in STF cloud,and the maximum value reached 1.23 g m?3. In this study, the LWC was bimodal in the PBL and generally varied little above it, which is inconsistent with in-situ observations of liquid-phase clouds in relatively clear regions ( Zhao et al., 2019 ). In the high LWC regions above 1500 m, the Nc varied little, the Na decreased, and the cloud Da as well as the aerosol Re increased slightly.

A high-level Na of 1853 cm?3was sampled by the PCASP near the surface from August in the Hebei region, which is lower than that in autumn and winter ( Liu et al., 2009 ; Zhao et al., 2018 ). In contrast to the Na, the vertical variation in the cloud droplet Re was large. Besides,a seesaw-type correlation was found between the cloud droplet Re and aerosol Na in high-Na regions. The LWC, mainly controlled by the Nc and cloud Da, responded little to the reduction in aerosol concentration.

All clouds studied in this paper were pure liquid-phase clouds and dominated by spherical particles. The cloud top was approximately 3 km in height, judging from radar echo. Horizontal detection measurements at various levels were performed from high to low. High radar echo reflectivity corresponded to large FCDP particle concentrations and small aerosol particle concentrations, whereas low radar echo reflectivity corresponded to small FCDP particle concentrations and large aerosol particle concentrations.

Funding

This research was funded by the National Key Research and Development Program of China [grant number 2017YFC1501405], the National Natural Science Foundation of China [grant numbers 41975180,41705119, and 41575131], and the National Center of Meteorology,Abu Dhabi, UAE (UAE Research Program for Rain Enhancement Science).