Using a recognition model of atmospheric gravity waves (AGWs), we identified 519 AGW events from the OH airglow images observed at the Dandong and Lhasa stations from 2015 to 2017. The 317 AGW events detected at the Dandong station have wavelengths ranging from 30 to 60 km, periods from 14 to 20 min, horizontal speeds from 30 to 60 m/s, and relative intensities from 0.4% to 0.6%, respectively. The parameters of 202 events recorded at the Lhasa station mainly vary within 15-35 km in horizontal wavelength, 4-6 min in period, 40-100 m/s in horizontal velocity, and 0.1%-0.3% in relative intensity. The occurrence rate peaks in winter and summer at Dandong and the peak in summer are absent at Lhasa because of the lack of convective weather. The seasonal propagation directions of the waves are influenced by both the wind field-filtering effect and the distribution of wave sources. In spring, because of the southeastward background wind field, fewer southeastward events are observed at the Dandong station. The situation at the Lhasa station is similar. In summer, both the Lhasa and Dandong stations are dominated by northeastward AGWs, which can be attributed to the southwestward wind. In autumn, ray-tracing results show that the events at Dandong mainly originate from wind shear, whereas the events at the Lhasa station are triggered by convective weather. The location of the wave sources determines the trend of the propagation directions at the Dandong and Lhasa stations in autumn. In winter, because of the eastward wind, more events are propagating to the southwest at the Dandong station.

Atmospheric gravity waves (AGWs) observed by the All-Sky Airglow Imager (ASAI) require accurate identification for the study of atmospheric coupling mechanisms and space weather prediction. However, the traditional manual screening methods and existing machine learning approaches do not meet the demands of practical station monitoring, which has significantly impeded climatological statistical research based on AGWs. Therefore, a real-time detection framework for ground-based airglow gravity waves that integrates transfer learning with adaptive image preprocessing has been proposed. By employing wavelength-adaptive median filtering and multiscale fusion, the framework effectively suppresses stellar noise while preserving weak gravity wave features. The model utilizes an EfficientNet-B3 (convolutional neural network) backbone enhanced with a deformable convolutional layer, trained via a two-stage strategy: A frozen phase prevents overfitting by locking the lower level feature extractor, and a fine-tuning phase optimizes the deformable convolution through cosine annealing and layered optimization. This approach improves both feature transfer efficiency and gravity wave detection sensitivity. The resulting lightweight model achieves 91.2% accuracy with millisecond-level inference speed (23 ms per frame).

Abstract. A compact solid etalon Fabry-Perot interferometer (SEFPI) with an effective aperture diameter of 70 mm has been developed for ground-based measurements of upper atmospheric winds and temperatures at an altitude of approximately 250 km. Unlike conventional FPIs employing air-spaced etalons, the SEFPI uses a solid etalon, enabling a more compact, lightweight, and cost-effective instrument design, while requiring stringent thermal stabilization. Laboratory validation using an acousto-optic frequency shifter (AOFS) to generate controlled Doppler shifts demonstrated the SEFPI’s high measurement accuracy, with an average wind uncertainty of 0.42 m/s and a root mean square error (RMSE) of 0.61 m/s. Field comparison validation was conducted via coincident ground-based measurements with two interferometry techniques. The first employed a Dual-Channel Optical Interferometer (DCOI), while the second utilized a well-established FPI employing a 100 mm diameter air-spaced etalon. The SEFPI measurements exhibited strong agreement with both instruments, yielding Pearson correlation coefficients exceeding 0.84. Statistical analysis yielded an overall mean wind difference of 1.88 m/s relative to the co-located DCOI instrument and 5.58 m/s relative to the conventional FPI located 502 km away. Temperature comparisons between the two FPIs revealed a correlation coefficient of 0.86 and an average difference of 62.38 K.

The zonal neutral wind is generally believed as the driving source of zonal drift of the equatorial plasma bubbles (EPBs). Comparing their correlations is crucial for understanding the zonal drift of EPBs. However, studies on their relationship by utilizing ground-based observational data are very limited, especially in the Chinese sector. In this study, we conducted a statistical comparison between the EPB zonal drift velocity estimated from an All-Sky Airglow Imager (ASAI) and the zonal neutral wind obtained from a Fabry-Perot Interferometer (FPI) and a Dual-Channel Optical Interferometer (DCOI) deployed at the Fuke (; , dip latitude similar to 9.7 degrees N) station over more than 1 yr, from June 2023 to December 2024. The results show that the zonal drift velocities of over 50% EPB events are generally consistent with the zonal neutral wind velocities. For the remaining cases, the EPB zonal drift velocity differs from the zonal neutral wind velocity, with more pronounced deviations observed around midnight. Before midnight, events with EPB zonal drift velocities lower than the wind velocity outnumber those with higher velocities. After midnight, this trend reverses, with higher EPB velocities becoming more frequent. Additionally, the differences between EPB zonal drift velocity and zonal neutral wind velocity exhibit seasonal variations and are modulated by geomagnetic activity. These results provide valuable insights for advancing our understanding of the patterns and physical mechanisms governing EPB zonal drift.

On the two consecutive days of February 26-27, 2014, the all-sky airglow imager in Qujing observed Equatorial Plasma Bubbles (EPBs) dissipating more rapidly than those that appeared before and after. While these rapidly dissipating EPBs were present, the Hainan COherent-scatter Phased Array Radar (HCOPAR) detected Field-Aligned Irregularities (FAIs) within them. These FAIs first descended and then rose. The vertical extent of the fluctuations exceeded 100 km. Observations from four ionosondes at different latitudes showed a consistent pattern over the two nights. The F-layer height initially ascended and then rapidly descended. Plasma from higher altitudes was compressed to lower altitudes, which resulted in a significant increase in peak density. As a result, Nighttime Ionosphere Enhancement (NIE) occurred. At first, the FAIs descended along with the F-layer. After a while, they began to rise again. The downward movement is believed to have accelerated the decay of EPBs in the NIE region.

Abstract. Severe thunderstorms drive vertical coupling through electrical processes associated with transient luminous events (TLEs) and dynamical processes associated with atmospheric gravity waves (AGWs). On 19 May 2022, an exceptionally intense thunderstorm outbreak occurred over the northern Bay of Bengal–Himalayan foothill region. The storm system produced more than 100 red sprites, 16 secondary gigantic jets, and at least four ghosts, accompanied by visible concentric airglow ripples. This event represented the first large-scale detection of a massive red sprite outbreak in China. From a dynamical perspective, concentric gravity waves (CGWs) were observed from the stratosphere to the thermosphere using coordinated satellite and ground-based airglow observations. AIRS measurements revealed concentric wave structures in the stratosphere above the convective source region, while ground-based airglow imagers detected CGWs in the OH, OI 557.7 nm, and OI 630.0 nm emission layers. CGWs in the mesosphere and lower thermosphere (MLT) propagated northeastward for more than 3000 km, suggesting propagation within favorable MLT ducting structures. In contrast, thermospheric CGWs observed in the OI 630.0 nm emission exhibited horizontal phase speeds of 260–280 m s⁻¹. Reverse ray-tracing analysis indicates a preferred source altitude near 150 km, suggesting that these waves were likely secondary gravity waves generated by the dissipation and breaking of upward-propagating primary gravity waves. Their asymmetric occurrence is likely attributable to Doppler-induced dissipation associated with thermospheric background winds. These observations provide a rare example of simultaneous electrical and dynamical coupling from the troposphere to the thermosphere during an extreme convective event.

In this study, we investigate the source and dynamics of a mesoscale gravity wave (MGW) observed over northern China. On January 12, 2010, an OH airglow imager at the Xinglong station (40.2 degrees N, 117.4 degrees E) detected an MGW propagating from southwest to northeast, consistent with the background wind direction. The wave exhibited a horizontal wavelength of 125 +/- 7.6 km, an observed period of 25 +/- 3.2 min, and a phase speed of 83 +/- 12.4 m/s. The momentum flux and the energy FLux of the MGW were approximately 24.93 m(2)/s(2) and 1.08 x 10(-5) W/m(2), respectively, from the airglow imaging observation. During propagation, wave breaking generated secondary ripples with wavelengths of 5-12 km. These ripples were likely caused by wind shear, as measured by the Doppler meteor radar at Shisanling (40.3 degrees N, 116.2 degrees E). According to OH emission profiles from the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) instrument on board the Thermosphere Ionosphere Mesosphere Energetics and Dynamics (TIMED) satellite, the height of the OH airglow layer was similar to 81 km during the MGW propagation event. A separate northwestward-propagating small-scale gravity wave with a wavelength of similar to 35 km was also observed. The backward ray-tracing analysis conducted with European Centre for Medium-Range Weather Forecasts (ECMWF) ERA5 reanalysis data indicated that the jet system near the Tibetan Plateau served as the source for the MGW.

This study presents novel observations of a mid-latitude red aurora event at 44 degrees MLAT over Northern China during the 5 November 2023 geomagnetic storm, using simultaneous observations from ground-based instruments (630 nm all-sky imager (ASI) and ionosonde) at Mohe station (53.5 degrees N, 122.4 degrees E; MLAT = 44.32 degrees N) and space-based satellites (Swarm and Defense Meteorological Satellite Program F17 and F18). The auroras displayed unusual dynamics, appearing north of the ASI field of view (FOV) at 1000 UT, disappearing at 1300 UT, then reappearing at 1400 UT with equatorward expansion (similar to 62 m/s) to reach 51 degrees N (42 degrees MLAT/2.3 L) and lasting similar to 8 hr. Analysis indicated that the auroral characteristics resembled stable auroral red arc (SARc), including thermal excitation (electron temperatures >3,000 K) and ionogram features (weak spread-F occurrence and reduced critical frequency). The auroras exhibited distinctive yet previously unreported properties of SARc, prompting their definition as SARc-like events. Specifically, the auroral emissions demonstrated clear additional contributions from ion precipitation (hundreds eV to tens keV ions) alongside thermal excitation of the typical SAR theory, and they were not fixed within the ionospheric trough unlike a narrow band of typical SARc pattern. The ionospheric trough at its equatorward edge exhibited synchronous equatorward motion. Concurrently, ionosonde echoes weakened and disappeared as the aurora and trough passed overhead, suggesting density depletion effects. Both the trough dynamics and ion precipitation likely drove electron temperature enhancements. These findings reveal new complexities in mid-latitude auroral physics requiring further investigation.

The three-dimensional spectral analysis method was applied to airglow data from September 2023 to August 2024 derived from an OH airglow imager located at the Hejing station (42.79 degrees N, 83.73 degrees E) to study the propagation characteristics of gravity waves (GWs) over Northwest China. We found that obvious seasonal variations occur in the propagation of GWs. In spring, GWs mainly propagate in the northeast direction. In summer and autumn, GWs mainly propagate in the north direction. However, GWs mainly propagate in the south direction in winter. The direction of GW propagation in the zonal direction is controlled by the wind-filtering effect, whereas the north-south meridional direction is mainly determined by the location of the wave source. We found that the average energy spectrum exhibits a 10%-20% higher intensity in summer and winter compared with spring and autumn. For the first time, we report the seasonal variation characteristics of GWs over the inland areas of Northwest China, which is of great significance for understanding the regional distribution characteristics of GWs.

Scattered airglow emissions in the lower atmosphere can bias ground-based interferometer observations of thermospheric winds, particularly when airglow brightness becomes spatially uneven due to auroras. During two geomagnetic storms with visible auroras on 10 May and 10 October 2024, the Doppler Asymmetric Spatial Heterodyne (DASH) and Fabry-Perot (FP) interferometers concurrently detected atypical winds at Siziwang (SIZW, 41.83° N, 111.93° E), suspected to be caused by scattering. These atypical winds, characterized by horizontal differences exceeding 400 m s−1 between opposite cardinal directions (N-S or E-W) and downwelling exceeding 100 m s−1, showed a strong temporal association with airglow brightness. By modelling the transmission of scattered airglow emissions, we calculate post-scattering wind speeds as the initial wind speeds weighted by both scattered and direct intensities. With fixed initial winds (100 m s−1 westward, 400 m s−1 southward, zero vertical wind), the simulation reproduces horizontal differences of approximately 400 m s−1 on 10 May and 100 m s−1 on 10 October, both capturing the temporal characteristics of the atypical winds. The simulation shows that scattering-induced biases on line-of-sight speed take their sign from the brighter region, while their magnitude varies directionally with the angle to that region: at 45° elevation, biases 135–180° azimuth away exceed those in the brighter region by more than 10 times. Limited by uncertainties in airglow images and optical depth of model inputs, the simulation incurs numerical errors of roughly 75 % during some periods. Effective correction of the scattering impact will require improved accuracy of model inputs in the future.

Winds in the mesopause region are key to understanding atmospheric dynamics. This study presents a novel ground-based Asymmetric Spatial Heterodyne Spectroscopy (ASHS) system designed to measure these winds by observing the nighttime green line airglow of atomic oxygen. The system's configuration, thermal behavior, and calibration procedures are detailed. The as-built configuration meets the performance expectations established during the design phase, enabling wind measurements with an accuracy better than 2 m s−1. Field observations from Mohe Station (53.5° N, 122.3° E) during geomagnetically quiet conditions demonstrate good agreement with co-located LiDAR data, validating the ASHS system's capability to derive mesopause winds from interferograms.

The equatorial plasma bubble (EPB) evolutionary behavior in the Chinese sector remains poorly understood due to insufficient long-term statistical characterization. Leveraging prior machine learning frameworks, this work presents the first statistical profile of EPB evolution over an entire solar cycle (2012-2022) in southern China, using airglow observations from Qujing, China (geographical: 25 degrees N, 104 degrees E; geomagnetic: 15.1 degrees N, 176.7 degrees E). The results reveal that: the relative occurrence frequency (ROF) peaks at similar to 22:15 local time (LT) and exhibits equinoctial asymmetry, with spring and autumn showing double pre-midnight peaks (22:15 and 23:30 LT), while winter displays a single peak (22:15 LT). The peak activities of the area within the observation field of view (FOV) and poleward magnetic latitude extension (PMLE) exhibit synchronous timing, predominantly occurring at 21:00, 23:00, 01:00, 03:30, and 05:00 LT. Before 02:00 LT, area in FOV and PMLE show seasonal variations: spring > autumn > winter. The zonal drift velocity (ZDV) peaks at 22:15 LT (primary), 01:45 LT (secondary), 03:45 LT (tertiary), and 05:45 LT (quaternary). Higher area in FOV, PMLE, and ZDV occur during high solar activity. Geomagnetic disturbances weakly increase ROF but suppress ZDV, PMLE, and area in FOV. Notably, our results show that the ROF near 00:00 LT drops sharply first and then rises suddenly, which persists regardless of statistical methodology. The correlations among PMLE, area in FOV, and speed exhibit seasonal variations, and are influenced by the intensity of solar and geomagnetic activity. These findings provide critical insights into EPB evolution in Chinese sector.

We developed a deep convolutional neural network‐based program code that automatically detects equatorial plasma bubbles (EPBs), segments EPB morphologies, and then extracts their features (including EPBs' northernmost foot latitudes and zonal drift velocities) from OI 630 nm airglow images. From 2012 to 2022, all‐sky airglow images from Qujing Station, China (geographic: 25°N, 104°E; geomagnetic: 15.1°N, 176.7°E), were manually labeled as EPBs or non‐EPBs. Some images showing typical EPB morphologies also had their contours annotated. This created two unique data sets which are both suitable for supervised learning models. Based on the above data sets, numerous experiments were conducted to train the EPB recognition model (EPB‐RM) and the EPB morphology segmentation model (EPB‐MSM). The results on the test set indicate that ResNet18+CBAM model can perform the best in recognizing EPB (precision: 99%, recall: 91%), and all models perform better during geomagnetically quiet periods and high solar activity periods (F10.7 value >140) than during geomagnetically disturbed periods (Dst index ≤ −30 nT or Kp index ≥3) and low solar activity periods. The EPB‐MSM based on Deeplabv3plus had the highest accuracy (88.2%) for extracting the EPB morphology. Statistical features extracted using the EPB feature extraction program (EPB‐FEP) were highly consistent with those extracted manually. The test results verified that machine learning is an excellent method for automatically detecting and extracting EPB characteristics. Our study provides a convenient tool for analyzing massive EPB images.

The responses of thermospheric winds at mid and low latitudes to the intense geomagnetic storm of November 3-4, 2021, are examined using multi-technique observations and simulations with the Thermosphere Ionosphere Electrodynamics General Circulation Model (TIEGCM). During the storm, the Kp index reaches 8 and the DST index reaches -105 nT. This study provides detailed profiles of thermospheric winds, especially the nighttime winds measured by the Michelson Interferometer for Global High-Resolution Thermospheric Imaging (MIGHTI). Significant nighttime wind disturbances are shown at latitudes between 30 degrees N and 42 degrees N, and there is a reversal structure in the zonal wind disturbances: westward below 250 km and eastward above it. Large westward and poleward nighttime disturbances emerge between 220 and 290 km at latitudes between 30 degrees N and 42 degrees N, coinciding with the minimum DST index. The maximum nighttime westward disturbance is -249 m/s around 220 km, and poleward is 193 m/s around 280 km. Daytime zonal wind disturbances are mainly eastward at latitudes between 0 degrees and 20 degrees N, except for latitudes between 15 degrees S and 0 degrees, where disturbances shift from westward to eastward as energy is injected. Significant daytime equatorward wind disturbances of -252 m/s at 160 km are shown as the DST index reaches the minimum between 10 degrees N and 20 degrees N. TIEGCM simulations underestimate MIGHTI after 07:00 Universal Time but are consistent with ground-based wind measurements.

A Dual-Channel Optical Interferometer (DCOI) based on Asymmetric Spatial Heterodyne Spectroscopy technique has been deployed at 11 stations by Phase II of the Chinese Meridian Project, establishing a ground-based network for routine neutral wind measurement in upper mesosphere and thermosphere over China. Neutral winds measured by the DCOI network in mesopause region were compared with average-weighted neutral winds from co-located meteor radar and sodium (Na) lidar in three stations. The DCOI winds are in a good agreement with meteor radar and Na lidar observations and the largest correlation coefficient reaches up to about 0.92. These results validate DCOI as a robust instrument for wind measurements in the mesopause region. The operational DCOI network will provide unprecedented spatiotemporal coverage of wind measurements in the mesopause for studying dynamic processes.

Three groups of intense concentric gravity waves (CGWs) lasting over 10 h were observed by an airglow imager at the Southern Space Observatory (SSO) in S & atilde;o Martinho da Serra (29.44 degrees S, 53.82 degrees W) in southern Brazil on 17-18 September 2023. These CGW events were simultaneously captured by spaceborne instruments, including the Atmospheric Infrared Sounder (AIRS) aboard Aqua, the Visible Infrared Imaging Radiometer Suite (VIIRS) on board Suomi NPP, and the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) instrument operating on the Thermosphere-Ionosphere-Mesosphere Energetics and Dynamics (TIMED) satellite. The CGWs caused significant airglow radiation perturbations exceeding 24 %, and the distance of the wave centre movement exceeded 400 km. These CGW events were caused by fast-moving deep convection observed by the Geostationary Operational Environmental Satellite-16 (GOES-16). The weaker background wind field during the spring season transition provides the necessary conditions for CGWs to propagate from the lower atmosphere to the mesopause region. The 630 nm emission images were significantly contaminated by specific OH emission bands. The same CGW event was observed propagating from the OH airglow layer (similar to 87 km) to the thermospheric OI 630.0 nm airglow layer (similar to 250 km). The asymmetric propagation of CGWs in the thermosphere may be due to the vertical wavelength changes caused by the Doppler-shifting effect of the background wind field. This multilayer ground-based and satellite joint detection of CGWs offers an excellent perspective for examining the coupling of various atmospheric layers.

A signal-tracking Fabry-Perot interferometer (STFPI) has been successfully developed using a frequency-stabilized He-Ne laser as a pilot signal for real-time capturing of system parameters simultaneously during nightglow observations. Wind velocities were derived in laboratory experiments using a novel piloted deconvolution technique from non-contaminating interference fringes generated from the He-Ne laser and a Neon calibration lamp emission at 630.479 nm. The piloted measurement method substantially improves measurement accuracy and eliminates sensitivity to laser calibration interval time, maintaining a consistent system uncertainty of 1.07 m/s in wind measurements. It also effectively suppresses the influence of instrumental drift. Under high-precision thermal control of an etalon over a 24-hour experiment period, the conventional measurement has an instrumental drift in wind velocity of 61.09 m/s, whereas the piloted approach consistently corrects this drift to below 5.25 m/s. These results confirm the STFPI's capability to enhance the stability and reliability of wind measurements under variable thermal conditions, supporting its application in long-term optical observations of faint nightglow emissions.

Abstract. Water-soluble components have significant contribution to the oxidative potential (OP) of atmospheric fine particles, while our understanding of their relationship is still limited. In this study, the water-soluble OP levels in wintertime PM2.5 in the south and north of Beijing, representing the difference in sources, were measured with dithiothreitol (DTT) assay. The volume normalized DTT (DTTv) in the north (3.5 ± 1.2 nmol min-1 m-3) was comparable to that in the south (3.9 ± 0.9 nmol min-1 m-3), while the mass normalized DTT (DTTm) in the north (65.3 ± 27.6 pmol min-1 μg-3) was almost twice that in the south (36.1 ± 14.5 pmol min-1 μg-3). In both the south and north of Beijing, DTTv was better correlated with soluble elements instead of total elements. In the north, soluble elements (mainly Mn, Co, Ni, Zn, As, Cd and Pb) and water-soluble organic compounds, especially light-absorbing compounds (also known as brown carbon), had positive correlations with DTTv. However, in the south, the DTTv was mainly related to soluble As, Fe and Pb. The sources of DTTv were further resolved using the positive matrix factorization (PMF) model. Traffic-related emissions (39.1 %) and biomass burning (25.2 %) were the main sources of DTTv in the south, and traffic-related emissions (> 50 %) contributed the most of DTTv in the north. Our results indicate that vehicle emission was the important contributor to OP in Beijing ambient PM2.5 and suggest that more study is needed to understand the intrinsic relationship between OP and light absorbing organic compounds.

The Phase Ⅱ of Chinese Meridian Project will deploy 10 meteor radar systems in China. To promote the advancement of domestic ground-based observation technologies, the project managing headquarters are engaged in a special campaign for the localization of meteor radar production. To ensure that the domestically manufactured product meets the requirements, prototype testing was carried out, including a technique indicator test and data quality evaluation. According to the results of the field technique indicator test, the prototype fully met the requirements. Further, the data quality was evaluated using the EMDR meteor radar data as a reference. The main comparison parameters included valid meteor count, space-time distribution of meteor number, height distribution of diffusion coefficient, and distribution of wind field at varying height and time. This study mainly explains data quality evaluation results, revealing some basic characteristics and laws of meteor radar observation, in an attempt to testify the capability of the prototype and provide a reference for future meteor radar data quality evaluation.

The concurrent observation of the Hainan COherent Scatter Phased Array Radar (HCOPAR) and the 630 nm all‐sky airglow imager was carried out on the night of 13 April 2015 in low‐latitudes of China. The plasma bubble recorded by the airglow imager and the Field‐Aligned Irregularities (FAIs) observed by the HCOPAR kept a good consistency. At the beginning, the FAIs were found to fill the entire plasma bubble, and they were drifting eastward together. During the dissipation process, the plasma bubble became wider and wider in zonal direction, while the airglow gray gradient on the eastern and western walls of the bubble had no much changes. Simultaneously, the echo Signal‐to‐Noise Ratio (SNR) of the irregularities in the bubble became lower and lower. The SNR at the depletion center decreased fastest, and the SNR of the FAIs on the western wall of the bubble fell more slowly than those in other parts. Polarization electric field in the bubble is suggested to have influence on the distribution of the FAIs. Gradient drift instability on the western wall supported the FAIs to live longer, and the FAIs on the eastern wall, namely on the front edge of the bubble, dissipated more rapidly.