Dongsong Sun

Stratospheric airship navigation requires accurate wind field measurements at a ~20 km altitude, where low pressure and density limit the effectiveness of conventional wind sensors. To address this, we present a compact direct-detection Rayleigh Doppler wind lidar based on the molecular double-edge technique. The system utilizes a 532 nm fiber-coupled pulsed laser (0.5 W, 5 ns) and a fixed-cavity dual-channel Fabry–Perot etalon as the frequency discriminator. A liquid crystal variable retarder (LCVR) combined with a polarization beam splitter (PBS) enables non-mechanical, high-speed beam switching between two orthogonal line-of-sight (LOS) directions for horizontal wind measurement. Systematic tests are performed in controlled wind fields within Mie-dominated and Rayleigh-dominated regimes. The lidar effectively captures the sharp radial velocity profiles at wind speeds up to 7.6 m/s. Comparative experiments with a reference anemometer show that the system delivers reliable performance at 0.48 m range resolution, with measurement uncertainty below 0.34 m/s. With its compact, lightweight, and high-precision design, the developed lidar demonstrates reliable wind measurement capability under laboratory conditions, indicating its potential for future deployment on stratospheric airships.

Accurate measurements of wind fields in the troposphere and stratosphere are essential for advancing atmospheric dynamics research, improving weather prediction, and supporting aerospace operations. However, a single Doppler lidar technique usually has limited capability to provide vertically extended wind profiles across both aerosol-rich lower altitudes and molecular-dominated higher altitudes. In this paper, we present a hybrid Doppler lidar system that combines a 355 nm scanning incoherent Rayleigh Doppler lidar with a 1550 nm coherent aerosol Doppler lidar for multi-scale wind field detection. The coherent Doppler lidar is used for boundary-layer wind retrievals, while the Rayleigh Doppler lidar, based on the double-edge technique, extends wind profiling from the upper boundary layer to approximately 40 km. Field deployments demonstrate continuous wind profiling from 50 m to 40 km, extending from the boundary layer to the stratosphere. Comparisons with radiosonde measurements show good agreement during the field campaigns, supporting the feasibility of this hybrid configuration for vertically extended wind profiling. The resulting high-resolution wind measurements across multiple atmospheric regions provide valuable data sources for studies of multi-scale circulation research, gravity wave dynamics, and climate-related atmospheric processes.

Large language models (LLMs) offer new possibilities for natural-language interaction with geospatial analysis systems, but their use in remote sensing instrument data analysis remains limited by weak execution control, poor reproducibility, and limited integration with domain-specific computation. The paper presents an agent for Incoherent Doppler wind LiDAR (ICDL) data analysis, named ICDL-Agent, a tool-augmented LLM framework for remote sensing instrument workflows. The system maps conversational user requests to executable analysis pipelines for wind retrieval, uncertainty estimation, visualization, and higher-level diagnostics through structured planning over a registry of domain-specific tools. To improve execution reliability, the system combines schema-constrained workflow generation, shared-state reuse of intermediate scientific products, and validation with bounded repair. In addition to supporting routine LiDAR processing, the framework can generate new tools when required and adapt to related analytical tasks through domain-aware guidance and procedural documentation. We evaluate the system on multiple atmospheric wind-observation datasets in China and show that it faithfully reproduces the refined Doppler wind-retrieval pipeline, achieving representative R²/MAE values of 0.52/3.73 m/s against ERA5 and 0.80/2.31 m/s against radiosonde observations, while supporting downstream analyses such as profile comparison, climatological interpretation, and gravity-wave diagnostics. More broadly, this study demonstrates how constrained LLM orchestration can support LiDAR researchers, remote-sensing instrument teams, and geospatial analysts seeking transparent, reproducible, and automated scientific data-processing workflows.

Measurements of atmosphere density in the upper thermosphere and exosphere are of great significance for studying space–atmosphere interactions. However, the region from 200 km to 1000 km has been a blind area for traditional ground-based active remote sensing techniques due to the limitation of facilities and the paucity of neutral atmosphere. To fulfill this gap, the University of Science and Technology of China is developing a powerful metastable helium resonance fluorescent lidar incorporating a 2 m aperture telescope, a high-energy 1083 nm pulsed laser, as well as a superconducting nanowire single-photon detector (SNSPD) with high quantum efficiency and low dark noise. The system is described in detail in this work. To evaluate the performance of the lidar system, numerical simulation is implemented. The results show that metastable helium density measurements can be achieved with a relative error of less than 20% above 370 km in winter and less than 200% in 270–460 km in summer, demonstrating the feasibility of metastable helium lidar.

The rapid development of artificial intelligence (AI) and deep learning has revolutionized the field of data analysis in recent years, including signal data acquired by remote sensors. Light Detection and Ranging (LiDAR) technology is widely used in atmospheric research for measuring various atmospheric parameters. Wind measurement using LiDAR data has traditionally relied on the spectral centroid (SC) algorithm. However, this approach has limitations in handling LiDAR data, particularly in low signal-to-noise ratio (SNR) regions. To overcome these limitations, this study leverages the capabilities of customized deep-learning techniques to achieve accurate wind profile reconstruction. The study uses datasets obtained from the European Centre for Medium Weather Forecasting (ECMWF) Reanalysis v5 (ERA5) and the mobile Incoherent Doppler LiDAR (ICDL) system constructed by the University of Science and Technology of China. We present a simulation-based approach for generating wind profiles from the statistical data and the associated theoretical calculations. Whereafter, our team constructed a convolutional neural network (CNN) model based on the U-Net architecture to replace the SC algorithm for LiDAR data post-processing. The CNN-generated results are evaluated and compared with the SC results and the ERA5 data. This study highlights the potential of deep learning-based techniques in atmospheric research and their ability to provide more accurate and reliable results.

The range of detection and accuracy of currently available Raman temperature LiDAR systems are primarily improved by increasing the energy or the aperture of the receiving telescope. However, this does not lead to a corresponding linear increase in the distance of detection and accuracy of the system. In this paper, the authors construct a simulation model and optimize its parameters to develop a Raman temperature LiDAR with low energy and a small aperture that has a maximum distance of detection of over 5 km during the day and over 10 km at night. The profile of the atmospheric temperature obtained through field tests was in good agreement with the results of a radiosonde. The maximum correlation between the Raman temperature LiDAR and the radiosonde was 0.94 at night and 0.81 during the day. The results showed that the proposed Raman temperature LiDAR, with low energy and a small aperture, can provide reliable data on the temperature in the troposphere throughout the day.

We report the development of a portable coherent Doppler lidar (CDL) system incorporating onboard field programmable gate array for real-time data processing. An ordinal statistic-based mean Doppler frequency estimation algorithm with high accuracy and low computational cost is proposed. The basic characteristics of the proposed method were specified with the use of mathematical modeling through the algorithm, realized as MATLAB computer code that includes parts simulating lidar signals and data processing. A comparison of the proposed method with established and well-used determination methods of the maximum Doppler frequency is accomplished. The ordinal statistics method is easy to implement and facilitates obtaining computational results swiftly. To demonstrate the performance of this lidar system, an intercomparison field campaign between lidar and ultrasonic anemometer is carried out. The excellent degree of correlation experimentally confirmed the feasibility of developing affordable, low-power consumption, compact, and robust CDL systems for various industrial and scientific applications.

The atmospheric molecular number density can be obtained from atmospheric temperature and pressure profiles and is a significant input parameter for the inversion of lidar measurements. When measurements of vertical profiles of temperature and pressure are not available, atmospheric models are typically considered a valid alternative option. This paper investigates the influence of different atmospheric models (forecast and reanalysis) on the retrieval of aerosol optical properties (extinction and backscatter coefficients) by applying Raman and elastic-only methods to lidar measurements, to assess their use in lidar data processing. In general, reanalyzes are more accurate than forecasts, but, typically, they are not delivered in time for allowing near-real-time lidar data analysis. However, near-real-time observation is crucial for real-time monitoring of the environment and meteorological studies. The forecast models used in the paper are provided by the Integrated Forecasting System operated by the European Centre for Medium-Range Weather Forecasts (IFS_ECMWF) and the Global Data Assimilation System (GDAS), whereas the reanalysis model is obtained from the fifth-generation European Centre for Medium-Range Weather Forecasts ReAnalysis v5 (ERA5). The lidar dataset consists of measurements collected from four European Aerosol Research Lidar Network (EARLINET) stations during two intensive measurement campaigns and includes more than 200 cases at wavelengths of 355 nm, 532 nm, and 1064 nm. We present and discuss the results and influence of the forecast and reanalysis models in terms of deviations of the derived aerosol optical properties. The results show that the mean relative deviation in molecular number density is always below ±3%, while larger deviations are shown in the derived aerosol optical properties, and the size of the deviation depends on the retrieval method together with the different wavelengths. In general, the aerosol extinction coefficient retrieval is more dependent on the model used than the aerosol backscatter retrievals are. The larger influence on the extinction retrieval is mainly related to the deviation in the gradient of the temperature profile provided by forecast and reanalysis models rather than the absolute deviation of the molecular number density. We found that deviations in extinction were within ±5%, with a probability of 83% at 355 nm and 60% at 532 nm. Moreover, for aerosol backscatter coefficient retrievals, different models can have a larger impact when the backscatter coefficient is retrieved with the elastic method than when the backscatter coefficient is calculated using the Raman method at both 355 nm and 532 nm. In addition, the atmospheric aerosol load can also influence the deviations in the aerosol extinction and backscatter coefficients, showing a larger impact under low aerosol loading scenarios.

Although polarization lidar technology has been widely used to detect depolarized particles, many difficulties still exist in measuring accurate values for the depolarization ratio. This ratio is basically related to the difference between the optical and the electronic gain of the lidar's channels, the polarization purity of the laser pulse, and the error in the alignment between the laser's polarization plane and the polarizer. We propose here an algorithm for the calibration and calculation of polarization lidars. Numerical simulations showed that the depolarization ratio profile retrieved from the proposed algorithm agrees well with the simulated profile. The proposed algorithm no longer requires the laser source to be linearly polarized. Instead, the light polarization can be elliptical or random. During the installation process, the polarization of the detector is no longer required to be parallel or perpendicular to the polarization direction of the laser source and can be done in any direction.