Jinqing Zhng, Hongbin Chen , Yunfei Du , Wenzheng Sho , Runping Zeng ,Keping Zhu , Yi Liu , Yuejin Xun

a Key Laboratory of Middle Atmosphere and Global Environment Observation, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China

b Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing, China

c College of Electronic Engineering, Chengdu University of Information Technology, Chengdu, China

d Jiangxi Xinyu Guoke Technology Co., Ltd, Xinyu, China

Keywords:Dropsonde technology Stratospheric balloon Development Atmospheric profile Tibetan Plateau

ABSTRACT To complement the atmospheric profile measurements under complex geographical environments and extreme weather conditions, a stratospheric balloon-based dropsonde technology, which is carried by a stratospheric balloon platform from the Earth’s surface to the upper troposphere and lower stratosphere (UTLS) to release the dropsonde for measurements, is independently developed and preliminarily assessed over the Tibetan Plateau(TP) in this study. The dropsonde system is mainly composed of the dropsonde chamber, dropsonde with a parachute, data receiving and communication antennas, dropsonde-releasing device, and GPS (Global Positioning System) modules. The dropsonde measurements can be sent in real time through satellite communication links and by radio signals to a data receiver at the ground control center for storage and processing. A total of eight dropsondes aboard the stratospheric balloon were successfully released during the TP campaign in 2020.A preliminary assessment was conducted based on a case comparison between the dropsonde and radiosonde measurements, which indicated that the dropsonde technology we developed can generally provide reasonable atmospheric profiles. However, further efforts are still required to improve the detection performance of the dropsonde sensors after long-term locating in the UTLS and to assess the accuracy and precision of the detection technology more carefully.

1. Introduction

The detection of vertical atmospheric profiles is significant for studying atmospheric processes, weather analysis, and climate forecasting. As a common and important in-situ observation technology, conventional meteorological sounding, which can provide detailed profiles of the atmospheric temperature, relative humidity (RH), air pressure, horizontal wind speed and direction from the surface up to ～35 km, is widely deployed to investigate the vertical structure of atmospheric thermodynamic characteristics ( MacCready, 1965 ; Seidel et al., 2010 ; Zhang et al.,2020a ). In China, there are around 120 meteorological sounding stations launching two radiosondes per day: one in the morning and one in the evening ( Guo et al., 2016 ; Zhang et al., 2018 ). Moreover, a global sounding observation network composed of more than 1000 observation stations around the world has provided radiosonde observations for several decades ( Durre et al., 2006 ; Wang and Wang, 2016 ).

Dropsonde technology has been proposed to complement the fixedpoint conventional meteorological sounding; i.e., a radiosonde attached to a parachute is released from a high-altitude observation platform(such as a balloon, airplane, etc.) to detect the vertical distribution of the atmospheric parameters ( Wang et al., 2009 ; Wick et al., 2018 ). Due to high adaptability and maneuverability, the dropsonde method was initially developed to study the vertical profiles of the kinematic and thermodynamic structures over vast seas and in severe weather conditions such as typhoons, hurricanes and tropical storms ( Franklin et al.,2003 ; Wu et al., 2007a ). Compared to radiosondes, dropsondes are much less available but have been successfully applied by foreign scholars for atmospheric research, such as in the African Monsoon Multidisciplinary Analysis from August to September in 2006 and the Concordiasi field experiment over Antarctica from September to December in 2010 ( Agustí-Panareda et al., 2010 ; Wang et al., 2013 ). However, to our knowledge,few studies have been conducted on the development of dropsonde technology in China (e.g., Zhang et al., 2019 ). Recent news media reported the application of unmanned aerial vehicle-based dropsonde measurements to Typhoon Senlake on August 2020 in the South China Sea, and further publication of the observation details and detection performance is expected in future literature.

Fig. 1. Schematic design (top panel) and working diagram (bottom panels) of the dropsonde technology.

The Tibetan Plateau (TP) is the highest (the average elevation is > 4 km above sea level: ASL) and broad (～2.5 ×10km) plateau in the world and substantially influences the regional weather and climate of the Northern Hemisphere through its unique thermal and dynamical forcings ( Wu et al., 2007b ; Yao et al., 2012 ). Moreover, the TP is located in the key area of the Asian summer monsoon (ASM) region, which acts as an efficient transport path of tropospheric atmospheric components and pollutants into the stratosphere ( Randel et al., 2010 ; Zhang et al.,2020b ). These pollutants, when transported into the stratosphere, are likely to generate an important radiative forcing on the regional and even global atmosphere ( Vernier et al., 2015 ; Yu et al., 2017 ). Due to the high elevation and harsh climate, few surface observation sites exist over the TP, especially in the central and northern TP, over which the atmospheric parameters are mainly derived from satellite observations( Zhao et al., 2020 ). To better comprehend the thermodynamic processes of pollutant transportation by the ASM into the stratosphere, we independently developed a stratospheric balloon-based dropsonde system to measure atmospheric profiles at high vertical resolutions and wide detection areas over the TP. This dropsonde system is also expected to complement the meteorological data under more complex environmental conditions, such as in “no-man’s land ”or sparsely populated areas(plateau and ocean), special terrain, and extreme weather conditions,thereby likely being a benefit to the study of atmospheric characteristics,accumulation of observation data, and evaluation of satellite detections and model simulations in such conditions.

The paper is organized as follows: The development of the stratospheric balloon-based dropsonde system is described in Section 2 .Section 3 presents the dropsonde experiment and a preliminary assessment of the dropsonde measurements over the TP, and a conclusion is given in Section 4 .

2. Development of the stratospheric balloon-based dropsonde technology

Fig. 1 shows the schematic design and working diagram of the dropsonde technology aboard the stratospheric balloon. The dropsonde system, which is mainly composed of a dropsonde with a parachute, data receiving and communication antennas, a dropsonde releasing device and a Global Positioning System (GPS) module, is carried by the stratospheric balloon platform from the surface to the upper troposphere and lower stratosphere (UTLS) to release the dropsonde from the chamber for measurements. To favour atmospheric contact and to provide reasonable atmospheric detection on the dropsonde falling path, the temperature and RH sensors of the dropsonde are oriented downwards and reach out from the bottom part of one lateral side of the dropsonde box.A pressure sensor is placed in the middle of the dropsonde box. The horizontal wind speed and direction are derived from the GPS information of the dropsonde. A total of eight dropsondes are deployed in the dropsonde chamber, which can be expanded according to the observation requirements. Two data receivers, one in the dropsonde chamber and one at the ground control center, are deployed to collect dropsonde measurements in real time. Due to the two-data-receiving-channels design, a data receiver can receive and process measurements from two dropsondes simultaneously falling in the sky.

The workflow of the dropsonde technology is specified as follows: After the stratospheric balloon taking offfrom the surface, the remote control command is sent in real time from the ground control center through a satellite communication link to the dropsonde releasing device, which controls the dropsonde release. Before releasing a dropsonde, we first send a dropping preparation instruction to trigger the self-check of the working status of the dropsonde. After completion of the self-check, a dropping command is sent to release the dropsonde by hanging it under a parachute. The atmospheric parameter profile is measured during the dropsonde falling process and is sent by radio signal to a data receiver in the dropsonde chamber. The atmospheric parameter profile is then further sent by a satellite communication link to a data receiver at the ground control center for storage and processing. In addition to the above data transmission line, the dropsonde measurements can also be directly sent by radio signal to the data receiver at the ground control center, provided that the dropsonde is released nearby to the ground control center (generally less than 100—200 km). More complete detection profiles are expected from the first transmission line compared to the latter due to radio signal obstruction of mountains.

3. Stratospheric balloon-based dropsonde experiment over the TP

3.1. Experimental description

The dropsonde system aboard the stratospheric balloon was launched at Qaidam (QDM; 37.74°N, 95.34°E; 3188 m ASL) on 23 August 2020 ( Fig. 2 ). Beijing time (BT) is used throughout the paper. QDM,which is located in northwestern Qinghai Province and at the northern margin of the TP, has an inland plateau desert climate. The stratospheric balloon was launched at 0520 BT and ascended to 21 km ASL after approximately one hour with a mean velocity of ～5 m s. It then flew at approximately 21 km until 2130 BT when the stratospheric balloon was cut and the flight platform started to descend while hanging under a parachute. The balloon flight lasted for roughly 16 hours in total and was approximately 100 km away from the QDM spot. All eight dropsondes deployed in the system were successfully released; among the eight dropsondes, one was released around the tropopause during the balloon ascent period and the other seven were released at approximately 21 km.

3.2. Preliminary assessment of the dropsonde measurements

To aid with comprehension of the detection performance of the dropsonde technology, a radiosonde was launched at QDM for comparison when we released the second dropsonde at 0650 BT. The radiosonde we used participated in the Eighth World Meteorological Organization International Radiosonde Comparison at Yangjiang, China, in 2010( Nash et al., 2010 ), and displayed a reasonable detection performance in the field campaign ( Xie et al., 2014 ). As an example, the measurements from this synchronous dropsonde and radiosonde are compared in this section to preliminarily assess the dropsonde detection performance( Fig. 3 ). Both the dropsonde and radiosonde provide measurements at a temporal resolution of 1 s in the sky. The relatively stable free lifting force from the latex ball guarantees the smooth vertical resolution of the radiosonde measurements at approximately 5 ms( Fig. 3 (a)).In contrast, a larger variation in the vertical resolution, which changes from ～14 msat 21 km to ～4 msat 5 km in altitude, is shown by the dropsonde measurements. The parachute used for the dropsonde is made of soft materials. The area of the stress surface may vary during the descent period due to the deformation and swing of the soft parachute by the horizontal wind, which will result in great variation in the vertical data resolution within a certain altitude range. However,owing to the air density increase when the dropsonde falls, the atmospheric buoyancy will increase and thereby induce a finer vertical data resolution with declining altitude.

The temperature profiles provided by the dropsonde and radiosonde were mostly in reasonable agreement, especially in the low and middle troposphere ( Fig. 3 (b)). However, a closer view revealed that a large discrepancy occurred in the UTLS (15—21 km), where a much higher temperature was recorded by the dropsonde than by the radiosonde. Except for the potential detection bias and time lag error of the temperature sensors, the discrepancy might be explained as follows. To ensure that the dropsondes can be powered on when we release them, the dropsondes are installed in an incubator to insulate from the cold atmospheric environment in the UTLS (approximately ? 60°C). This design will result in a higher temperature of the released dropsonde than its surrounding atmosphere, especially in the UTLS with thin air that is weakly conducive to heat transfer between the temperature sensor and the atmosphere. In contrast, the temperature detections from the dropsonde and radiosonde agree well in the low and middle troposphere (below 15 km), where the increased air density helps promote heat transfer.

Overall, the dropsonde-based RH profile captured the vertical moisture structure reasonably well and followed the general features seen in the radiosonde measurements ( Fig. 3 (c)). At a more detailed level,the dropsonde tended to detect higher RHs and placed the RH profile slightly higher in the atmosphere than the radiosonde, which was likely due to a large spatial variation in water vapour during the drift of the dropsonde and radiosonde. A consistent pressure profile was provided by the two kinds of soundings ( Fig. 3 (d)). Both the horizontal wind speed and direction bore a close resemblance between the dropsonde and radiosonde measurements ( Fig. 3 (e and f)), although slight discrepancies existed in the details, which were also likely associated with the spatial variation in the wind field and due in part to different vertical resolutions between the dropsonde and the radiosonde.

Fig. 2. (a) Eight dropsondes released on the stratospheric balloon flight path. The pink star denotes the location of the QDM site (37.74°N, 95.34°E; 3,188 m ASL).The thick line rising from the QDM surface indicates the flight path of the stratospheric balloon; the color of the line presents the flight time (Beijing time in hours on 23 August 2020, as shown by the top color bar) of the balloon whose ascent flight is primarily denoted by blue color and the level and descent flights are denoted by the other colors. The black rectangles show the dropsonde release spots on the balloon flight path, and the black fine lines represent the detection path of the dropsondes; the sequence number of the dropsonde is also noted, i.e., from D1 to D8. The gray fine line shows the ascending path of the radiosonde launched at QDM. The bottom color bar shows the ground elevation of the TP (ASL: m). (b) The altitude variation along the flight path of the balloon.

4. Conclusion and discussion

Conventional meteorological sounding has long been an important approach for providing detailed profiles of vertical atmospheric parameters across the world. For example, the radiosonde has been in routine operation at ～120 stations in China for over a decade. Even so,more observation techniques are still needed since current measurements are still insufficient in complex geographical environments and extreme weather conditions. In this study, a stratospheric balloon-based dropsonde technology was independently developed and preliminarily assessed in an atmospheric campaign over the TP in 2020.

A total of eight dropsondes aboard the stratospheric balloon were successfully released during the TP campaign. We preliminarily assessed the credibility of the dropsonde technology based on a case comparison with a synchronous radiosonde. The results indicated that the dropsonde technology presented here could generally provide reasonable atmospheric measurements and should have the potential to be deployed for meteorological observations and scientific applications under various complex environmental conditions. However, a much higher temperature tended to be recorded by the dropsonde than by the radiosonde in the UTLS (15—21 km), where the thin air is not conducive to heat transfer between the temperature sensor and the atmosphere. Further developments are required to improve the detection performance of the dropsonde sensors after being positioned for a long time in the UTLS with a low temperature and low pressure. In addition, further studies will focus on a more careful assessment of the detection accuracy and precision of dropsonde technology we developed. Satellite measurements and a surface-combined network that can cover the predicted drifting path of the dropsonde should help achieve this assessment, especially in the case of large drift of the stratospheric balloon.

Fig. 3. Comparisons of the (a) vertical data resolution, (b) atmospheric temperature, (c) RH, (d) pressure, (e) horizontal wind speed and (f) direction profiles provided by the dropsonde (red line) and radiosonde (blue line) synchronously launched at QDM.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was supported by the Strategic Priority Research Program of Chinese Academy of Sciences [grant number XDA17010101 ],the National Natural Science Foundation of China [grant number 41875183 ], and the National Key R&D Program of China [grant number 2017YFA0603504 ] .

Acknowledgement

The authors would like to acknowledge all the members for their contributions to the Qaidam atmospheric experiment in 2020. Special thanks are extended to the Aerospace Information Research Institute,Chinese Academy of Sciences, for providing and releasing the stratospheric balloon during the campaign.