CHEN Zeyu CHEN Hongbin XU Jiyao HUANG Kaiming XUE Xianghui HU Dingzhu CHEN Wen YANG Guotao TIAN Wenshou HU Yongyun XIA Yan

Advances in the Researches of the Middle and Upper Atmosphere in China

CHEN Zeyu1,6CHEN Hongbin1,6XU Jiyao2,6HUANG Kaiming3XUE Xianghui4HU Dingzhu5CHEN Wen1,6YANG Guotao2TIAN Wenshou7HU Yongyun8XIA Yan8

1 (100029) 2 (100190) 3 (430072) 4 (230026) 5 (210044) 6 (100049) 7 (730000) 8 (100871)

In this report we summarize the research results by Chinese scientists in 2018–2020. The focuses are placed on the researches of the middle and upper atmosphere, specifically the researches on atmospheric structure and composition, climate and chemistry-climate coupling and climate modelling, dynamics in particular those inducing the coupling of the atmospheric layers.

Middle and upper atmosphere, Structure and composition, Climate, Layer coupling

1 Structure and Composition in the Middle Atmosphere

Xia.[1]investigated the impact of tropical tropopause warming on the stratospheric water vapor using the Specified-Dynamics version of the NCAR Whole Atmosphere Community Climate Model. It is found that the tropical tropopause warming results in a strengthening of the Brewer-Dobson Circulation (BDC). The strengthening of BDC induced by a narrow warming of tropical tropopause within 12° latitude, which is much stronger in boreal winter than that in boreal summer, propagates more dry air from the tropical tropopause into the stratosphere and thus causes a reduction of water vapor in the middle stratosphere. On the contrary, the seasonal difference of the BDC strengthening is weaker in the experiment with a broader tropical tropopause war-ming within 25° latitude. The drying effect of the BDC is counteracted by the moistening effects of the tropical tropopause warming and methane oxidation. This leads to the moistening in both the lower and upper stratosphere. The results suggest the control of the stratospheric humidity by the tropical tropopause temperature could be significantly offset by the associated BDC changes.

Xia.[2]showed that stratospheric ozone- induced cloud radiative effects also play important roles in causing changes in Antarctic sea ice. Their simulations demonstrate that the recovery of the Antarctic Ozone Hole causes decreases in clouds over Southern Hemisphere (SH) high latitudes and increases in clouds over the SH extratropics. The decrease in clouds leads to a reduction in downward infrared radiation, especially in austral autumn. This results in the cooling of the Southern Ocean surface and increasing Antarctic sea ice. Surface cooling also involves ice-albedo feedback. Increasing sea ice reflects solar radiation and causes further cooling and more increases in Antarctic sea ice.

Liu and Fu[3]constructed an approximate month- to-month temperature change equation and extended it to a new form for decade-to-decade changes. Their result showed that at 100 hPa, the month-to-month Arctic temperature increment is a small term compared to the dynamical heating and diabatic heating, which are largely canceling terms with maximum magnitudes in November to April and October to March,respectively. However, it is not the case for their decadal changes and the decadal change of the Arctic current-month temperature compared to those of the regressed dynamical heating and radiative heating, where the current-month decadal cha-nges and the corresponding trends are approached except in March and a rough agreement exists between these trends and those reported in other studies. The dynamical plus diabatic heating term and the temperature increment, as well as their decadal changes, are roughly balanced during the annual oscillation. However, some departures exist in both cases because of the large deviations or uncertainties of relevant terms and also probably due to the quasi-geostrophic approximation and the eddy heat flux approximation of the dynamical heating, and a restricted condition of the eddy heat flux approximation is given at the end.

Based on measurements of the solar cycle by the Spectral Irradiance Monitor onboard the SORCE satellite, monthly ERA-Interim Reanalysis data, the radiative transfer scheme of the Beijing Climate Center (BCC-RAD) and a multiple linear regression model, Shi.[4]showed that during periods of strong solar activity, the solar shortwave heating anomaly from the climatology in the tropical upper stratosphere triggers a local warm anomaly and st-rong westerly winds in mid-latitudes, which streng-thens the upward propagation of planetary Wave 1 but prevents that of Wave 2. The enhanced westerly jet makes a slight adjustment to the propagation path of Wave 1, but prevents Wave 2 from propagating upward, decreases the dissipation of Wave 2 in the extratropical upper stratosphere and hence wea-kens the Brewer-Dobson circulation. The adiabatic heating term in relation to the Brewer-Dobson circulation shows anomalous warming in the tropical lower stratosphere and anomalous cooling in the mid-latitude upper stratosphere.

Hu.[5]found an out-of-phase relationship between the northern Hadley circulation (HC) extent (HCE) and Arctic stratospheric ozone (ASO) during boreal spring on interannual timescales during 1979–2014. Decreased (increased) ASO tends to result in a poleward (equatorward) shift of HCE by +0.67° (–0.45°) latitude per standard deviation of decreased (increased) ASO year. Observational ana-lysis and model simulations showed that increased ASO leads to an equatorward shift of HCE and subtropical moisteningweakened eddy momentum flux and decreased subtropical static stability, accompanied with negative eddy momentum flux divergence anomalies and northward meridional wind anomalies over the subtropics. Their results may help to understand the linkage between the Arctic and mid-latitudes, especially important for the subtropical precipitation and hydrological cycle.

Ma.[6]reported a Rayleigh and sodium lidar system, which was recently upgraded at the University of Science and Technology of China (USTC) in Hefei, China (31.5°N, 117°E). The lidar system has features of high temporal and vertical resolutions, a high signal-to-Noise Ratio (SNR), and mobility. Using the time-division and wavelength-division multiplexing methods, only one piece of the photomultiplier tube is used in the optical receiver, which makes the system compact and robust. Wideband filtering and narrowband filtering are both used in the lidar system to obtain high SNR data under city lights. The lidar system was established on 24 September 2016 and has run stably for 2 years. Meteor trail events that lasted for only a few seconds were extracted from the high resolution, high SNR observational data. The sodium observational data in 2017 were fitted annually and semiannually, and the results were similar to those obtained in previous stu-dies. The monthly average atmospheric temperature showed semiannual variations. Stratospheric aerosols were observed for two consecutive days during the observations.

Zou.[7]presented a general response of the high latitude mesospheric HO2and O3to several extremely large SPEs, using the Microwave Limb Sounder (MLS) observations onboard the Aura satellite. Through complicate ion chemistry, energetic protons that deposited in the atmosphere lead to the HO2enhancement of more than 0.2 ppbv (part per billion volume) above 0.2 hPa for 2 days. In a similar height range, the catalytic decomposition of O3is more than about 10% and the peak values of about >20% can last 1~3 days. The superposition analysis found that the O3depletion was delayed by 1 day relative to the HO2increment, which further indicates that HO2catalyzes the decomposition of O3.

Xun.[8]reported the first concurrent observations of thermospheric Na layers from two nearby lidar stations located at Yanqing (40.5°N, 116.0°E) and Pingquan (41.0°N, 118.7°E). From one year data set, they found four such layers, including an unprecedented one reaching 200 km with the highest density of 35 cm–3and fastest descending rate. While the main Na layers were comparable, in three nights, these thermospheric layers were observed only in one station (Yanqing), suggesting that these layers often occur locally with a horizontal scale less than about 250 km. They tabulate, compare, and discuss the principle characteristics of all the reported thermospheric layers.

Jiao.[9]presented the first simultaneous observation of mesopause sodium (Na) and potassium (K) layer by a Na-K lidar at the South Hemisphere site, S?o José dos Campos (23.1°S, 45.9°W). On 21 November 2016, sporadic layers in both Na and K layer occurred in main layer height with obvious descending variations with time, which seems like tidal induced. Notably, the peak K/Na ratio slowly increased with time. And Na layer and K layer showed different processes along with time with K density reaching its maximum 1 h later than that of Na.

Wang.[10]found a special behavior of the potassium layer in the mesopause region, by analyzing Yanqing lidar data. They called it as lower- triangle potassium layer, which appeared frequently in January. The frequent appearance of the lower- triangle potassium layer has made the average column density and peak density increase by 15.7% and 12.9%. It can be speculated that the increasing potassium atoms in the lower-triangle potassium layer were mainly converted from KO2, and partly from KOH.

Yang.[11]found a prominent Sporadic Sodium Layer (SSL), which is the largest one in 222 SSL events found from Haikou lidar data. The peak density of this large SSL was as high as 37087 cm–3. By comparing the corresponding data of the nearest ionosonde and VHF radar at Danzhou(19.5°N, 109.1°E), it was found that SSL is closely related to Es.

Based on the FPI observations and HWM14 and TIEGCM model results, Jiang.[12]studied the thermospheric horizontal winds at three FPI stations: Xinglong (40.2°N, 117.4°E), Kelan (38.7°N, 111.6°E) and Millstone Hill (42.6°N, 71.5°W). The results showed that the winds at Millstone Hill were more southward and more westward in four seasons com-pared to the other two stations; the directional reversal time of zonal winds at Millstone Hill was earlier than the other two stations. TIEGCM model was more consistent with FPI observations in the winter months compared to summer. Furthermore, TIEGCM model could in general replicate the observations at Millstone Hill. HWM14 model-data discrepancies mainly appeared in the winter zonal winds. Overall, among three stations, HWM14 predicted MH FPI observations best.

Liu.[13]studied the responses of neutral temperature in the lower thermosphere to the 2013 St. Patrick’s Day geomagnetic storm based on TIMED/SABER and AIM/SOFIE satellites observations. The results indicated that the temperature in the two hemispheres varied drastically due to the storm. The variation of the temperature depended on latitude, altitude, and the phase of the storm, and the maximum variation of the temperature could exceed 15 K. The temperature increase reached its peak 0.5~1.5 d (depending on latitude and height) after that theindex reached its peak.

Liu.[14]studied the responses of the multiday oscillations in thermospheric temperature to oscillations in10.7andindex by using the nighttime thermospheric temperature (about 250 km) measured by FPI at Xinglong (40.2°N, 117.4°E) station between 2010 and 2018. The results showed that the 27, 13.5, 9, and 7-day oscillations depended on solar phases. The 27-day oscillation was predominant during solar maximum and highly correlated with the10.7andindex. The 13.5, 9, and 7-day oscillations were important and highly correlated withindex during solar ascending phase.

In addition, Yang.[15]compared three me-thods (radius method, complete Fourier series description method and nonlinear regression fitting method) to derive neutral winds from ground-based FPI observation. The results showed that the non-linear regression fitting method was the best because more comprehensive fringe information was taken into account to derive the neutral winds by this method.

Xu.[16]simulated the impacts of super volcanoes on ozone depletion by using a transport model and a coupled chemistry-climate model, since strong volcanic activity can cause ozone depletion that might be severe enough to threaten the existence of life on Earth. In their experiments, the volcanic eruptions were the 1991 Mount Pinatubo eruption and a 100× Pinatubo size eruption. The results indicated that the percentage of global mean total column ozone depletion in the 2050 RCP8.5 100× Pinatubo scenario could reach approximately 6% compared to two years before the eruption and 6.4% in tropics. Another identical simulation, 100×Pina- tubo eruption only with natural source ozone depleting substances (ODSs), produced an ozone depletion of 2.5% compared to two years before the eruption, and with 4.4% loss in the tropics. Their model results suggested that the reduced ODSs and stratospheric cooling could lighten the ozone depletion after super volcanic eruption.

2 Climate and Modeling

Xia.[17]revisited the problem of the SH stratospheric warming in the recent decade. It is found that the SH high-latitude stratosphere continued warming in September and October over 2007–2017, but with very different spatial patterns. Multiple linear regression demonstrates that ozone increases play an important role in the SH high-latitude stratospheric warming in September and November, while the changes in the Brewer-Dobson circulation contributes little to the warming. This is different from the situation over 1979–2006 when the SH high-latitude stratospheric warming was mainly caused by the strengthening of the Brewer-Dobson circulation and the eastward shift of the warming center. Simulations forced with observed ozone changes over 2007–2017 shows warming trends, suggesting that the observed warming trends over 2007–2017 are at least partly due to ozone recovery. The warming trends due to ozone recovery have important implications for stra-tospheric, tropospheric, and surface climates on SH.

Li.[18]investigated the independent and joint impacts of ENSO Modoki and QBO on strato-spheric ozone in the Northern Hemisphere (NH) in winter. Their results showed that stratospheric ozone in the NH in winter increases during El Nino Modoki events but increases during La Nina Modoki, whereas increases during the easterly phase of QBO and decreases in the westerly phase of QBO. Further, they found that the EQBO enhances the effect of El Nino Modoki events on stratospheric ozone, but weakens the effect of La Nina Modoki, and vice versa for the WQBO.

Xie.[19]found that replacing the original specified Arctic Stratospheric Ozone (ASO) forcing with more accurate stratospheric ozone variations improves the simulated variations of global surface temperature in a climate model. The decreasing trend of stratospheric ozone may have enhanced the warming trend at high latitudes in the second half of the 20th century. Xie.[20, 21]and Ma.[22]further found that ASO changes have significant regional impacts.

Xie.[23]found that a warmer Indo-Pacific Warm Pool (IPWP) significantly dries the stratospheric water vapor by causing a broad cooling of the tropopause, and vice versa for a colder IPWP. They further found the frequency of ENSO Modoki events was higher from 1984 to 2000 than after 2000, the BD circulation anomalies related to central ENSO were stronger during 1984–2000, which caused ENSO Mo--doki events to have a greater effect on lower strato-spheric ozone before 2000 than eastern ENSO[24].

Zhang.[25]identified a Eurasia-North Ame-rica dipole mode in the total column ozone (TCO) over the Northern Hemisphere, showing negative and positive TCO anomaly centers over Eurasia and North America, respectively. The positive trend of this mode explains an enhanced TCO decline over the Eurasian continent in the past three decades, which is closely related to the polar vortex shift towards Eurasia. Moreover, they found that the positive Eurasia-North America dipole trend in late winter is likely to continue in the near future.

Zhang.[26]found that there are significant negative total column ozone (TCO) trends over the North Pacific and positive TCO trends over northwestern North America in winter during 1979–2015. The zonally asymmetric TCO trends are mainly contributed by dynamical processes related to the teleconnection pattern changes of the Cold Ocean - Warm Land (COWL) and the North Pacific (NP). By contrast, chemical processes make a relatively smaller contribution to the zonally asymmetric TCO trends.

Shangguan.[27]analyzed the variability and trends of temperature and ozone in the UTLS for the period of 2002–2017. A significant warming of 0.2~0.3 K per decade is found in most areas of the troposphere while the stratospheric temperature decreases at a rate of 0.1~0.3 K per decade. They also found the temperature increase in the troposphere, as well as ozone decrease in the NH stratosphere, is mainly connected to the increase of SST and subsequent changes of atmospheric circulations.

Xiao.[28]investigated the impact of increasing surface emissions of nitrogen oxide (NO) in East Asia on ozone and temperature in the UTLS region. Their results showed that in summer the south Asia anticyclone can transport the NOin the East Asian UTLS region to the low latitudes and result in an increase in the low-latitude ozone concentration, leading to warming in the UTLS region in winter, while the intensified cyclic ozone depleting in the mid-latitudes results in the mid-latitude ozone concentration decreasing, leading to cooling in the UTLS region in winter.

Yi.[29]estimated neutral mesospheric densities at low latitude have been derived from April 2011 to December 2014 using data from the Kunming meteor radar in China (25.6°N, 103.8°E). The daily mean density at 90 km was estimated using the ambipolar diffusion coefficients from the meteor radar and temperature from the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) instrument. The seasonal variations of the meteor radar derived density are consistent with the density from the Mass Spectrometer and Incoherent Scatter (MSIS) model, show a dominant annual variation, with a maximum during winter, and a minimum during summer. In addition, a comparison of the ambipolar diffusion coefficient and peak height observed simultaneously by two co-located meteor radars indicates that the relative errors of the daily mean ambipolar diffusion coefficient and peak height should be less than 5% and 6%, respectively; the absolute error of the peak height is less than 0.2 km.

Wu.[30]analyzed a long-term simulation of the Whole Atmosphere Community Climate Model with the chemistry of three metals (Na, K, and Fe) to investigate the response of the meteoric metal layers in the mesosphere and lower thermosphere regions to the 27-day solar rotational cycle. This is the first time that the solar-driven 27-day variation of the metal layers in the MLT region has been identified. The correlation between variability in the metal layers and solar 27-day forcing during different phases of the solar 11-year cycle reveals that the response in the metal layers is much stronger during solar maximum. The altitude-dependent correlation and sensitivity of the metal layers to the solar spectral irradiance demonstrate that there is a significant increase in sensitivity to solar rotational cycle with increasing altitude. Above 100 km, the sensitivity of the metals to changes of 10% in the solar spectral irradiance at Lyman-alpha is estimated to be –5%. A similar response is seen in Na layer measurements made by the Optical Spectrograph and InfraRed Imaging System instrument on the Odin satellite.

Yi.[31]presented observations of a quasi- 90-day oscillation in the Mesosphere and Lower Thermosphere (MLT) region from April 2011 to December 2014. There is clear evidence of a quasi- 90-day oscillation in temperatures obtained from the Kunming meteor radar (25.6°N, 103.8°E) and the SABER, as well as in wind observed by the Kunming meteor radar. The amplitudes and phases of the quasi-90-day oscillation in the SABER temperature show a feature similar to that of upward-propagated diurnal tides, which have a vertical wavelength of about 20 km above 70 km. Similar to the quasi- 90-day oscillation in temperature, a 90-day variability of ozone (O3) is also present in the MLT region and is considered to be driven by a similar variability in the upwardly-propagated diurnal tides generated in the lower atmosphere. Moreover, the 90-day var-iability in the absorption of ultraviolet (UV) radiation by daytime O3in the MLT region is an in situ source of the quasi-90-day oscillation in the MLT temperature.

Yi.[32]reported a climatology of the global mesopause relative density estimated using multiyear observations from meteor radars located from high to low-latitude regions. The seasonal variations of the southern polar mesopause relative density are mainly dominated by an Annual Oscillation (AO). The mesopause relative densities at high latitudes and high mid-latitudes in the Northern Hemisphere show mainly an AO and a relatively weak Semiannual Oscillation (SAO). The SAO is evident in the Nor-thern Hemisphere, especially at high latitudes, which is comparable to the AO amplitudes. The mesopause relative densities at lower mid-latitudes and low latitudes show mainly an AO. These observations indicate that the mesopause relative densities over the southern and northern high latitudes exhibit a clear seasonal asymmetry. The maxima of the yearly variations in the mesopause relative densities display a clear latitudinal variation across the spring equinox as the latitude decreases; these latitudinal variation characteristics may be related to latitudinal changes influenced by gravity wave forcing.

Rao.[33]revealed that the ratio of the ensemble members that forecast the zonal wind reversal with a 5-day delay allowed (hit ratio) is higher for SSW events with a small decrease in the zonal mean zonal winds (moderate SSWs) than for events with a large decrease in the zonal mean zonal winds (radical SSWs) in hindcasts initialized around 1 (7 day) and 2 (14 day) weeks in advance. The underestimated cumulative eddy heat flux associated with weak wave activities accounts for the weaker-than-observed deceleration of westerlies. The preexisting extratro-pical wave patterns are satisfactorily forecast in 14 day for most (9/12) cases, and the wave phase bias is reasonably small for those cases. After the climatology bias is deducted from the hindcasts, an increase in the hit ratio can be identified for moderate SSW events as the evolutions of zonal winds are improved. Following the error correction by remapping the zonal wind probability distribution function in the forecast system to the reanalysis, the SSW hit ratios increase in the 7 day (43% to 57%) and/or 14 day (11% to 21%) initializations. Rao.[34]also found when the hindcasts are initiated less than two weeks before SSW onset, BCC_CSM and ECMWF show comparable predictive skill in terms of the temporal evolution of the stratospheric circumpolar westerlies and polar temperature up to 30 days after SSW onset. However, with earlier hindcast initialization, the predictive skill of BCC_CSM gradually decreases, and the reproduced maximum circulation anomalies in the hindcasts initiated four weeks before SSW onset replicate only 10% of the circulation anomaly intensities in observations. The earliest successful prediction of the breakdown of the stratospheric polar vortex accompanying SSW onset for BCC_CSM (ECMWF) is the hindcast initiated two (three) weeks earlier.

Yu.[35]evaluated the prediction skill for the stratospheric mass circulation variability in winter (November to March) in the Climate Forecast System, version 2 (CFSv2), from 2011 to 2018. Three strato-spheric mass circulation indices measuring meridio-nal mass transport into the polar stratosphere by the total flow (ST60N), wavenumber-1 (ST60N_W1), and wavenumber-2 waves (ST60N_W2) are consi-dered. Systematic forecast bias is found in both the 7-year averaged winter mean and seasonal cycle, which is tied to the overestimation of damping in amplitude and westward tilting variations of total waves and difficulties in forecasting the exact con--tributions from different spatial scales of waves. The intra-seasonal variations of stratospheric mass circu--lation indices, with the systematic forecast bias cor--rected, can be modestly predicted at a forecast lead time of about 20 days, in terms of both the anomaly value and timing of negative and positive peak events. The 20-day prediction limit of the stratospheric mass circulation indices is mainly due to the 2-week limit of the CFSv2 model in predicting the variability of anomalous wave tilt angle, whereas the prediction limit of the wave amplitude anomaly can exceed 50 days.

3 Dynamics in the Middle Atmosphere

3.1 Meteorological Process

Hu and Ren[36]reported a significant linkage bet-ween the seasonal timing of boreal spring strato-spheric final warming (SFW) events and the onset of the Indian summer monsoon (ISM). The leading singular vector decomposition pattern reveals a significant, coupled interannual variation between the SFW-related circulation in the lower stratosphere and the ISM-related circulation in the lower troposphere, objectively confirming the intimate relationship between SFWs and the ISM. Associated with a late SFW, the stratospheric polar vortex and the polar jet are anomalously stronger in April to early May, which is coupled with positive anomalies of the Northern Annular Mode (NAM) and the Arctic Oscillation (AO) in the troposphere. These tropospheric NAM/AO anomalies act to pass the extratropical anomaly signals to western central Asiaa NAM-/AO-related Rossby wave train in the upper troposphere, which is initiated over the North Atlantic jet exit region and extends across Eurasia. This results in an anomalous upper tropospheric anticyclone accompanied by anomalous descent over western central Asia that in turn warms the in situ air column and results in an enhanced meridional gradient of tropospheric temperature over the land to the north of the Arabian Sea and the Indian Ocean and therefore an early onset of the ISM.

Yu.[37]revealed that SSW events correspond to a large-amplitude or long-lasting subset of pulse- like, anomalously strong, stratospheric mass circulation events. The anomalously strong, stratospheric mass circulation events (referred to as PULSE events) occur more than nine times in an average winter. The displacement versus split types of SSWs tend to correspond to the Wavenumber 1 versus Wavenumber 2 types of PULSEs, though the relationship between split-type SSWs and Wavenumber-2-type PULSEs is weaker. Like SSW events, PULSEs also have a close relationship with CAOs. The robust relationship with CAOs still holds for the PULSE events not accompanied by SSW events. More than 70% of CAOs in the 37 winters occur in the week before and after a PULSE event, with a false alarm rate of CAO occurrence of about 25.7%. SSW events, however, are associated with only about 5.7% of CAOs, with a false alarm rate of 21.7%. Therefore, the linkage between individual continental-scale CAOs and PULSE events represents a more generalized relationship between the stratospheric circulation anomalies and surface weather.

Hu.[38]reported a statistically significant relationship between the North Pacific Gyre Oscillation (NPGO) and stratospheric final warming (SFWOD) in the Northern Hemisphere in two sub- periods (1951–1978 and 1979–2015),, in the first (second) sub-period, the NPGO is negatively (positively) linked with SFWOD. During 1951–1978, positive NPGO years tend to strengthen the Pacific- North America (PNA) pattern in the mid-troposphere in boreal winter. The strengthened PNA pattern in February leads to strong planetary wave activity in the extratropical stratosphere from late February to March and causes the early onset of SFW in early April. By contrast, a strengthened Western Pacific pattern from January to early February in negative NPGO years causes a burst of planetary waves in both the troposphere and extratropical stratosphere from late January to mid-February and results in more winter stratospheric sudden warming events, which, in turn, leads to a dormant spring and a late onset of SFW in late April. During 1979–2015, positive (negative) NPGO years strongly strengthen (weaken) the mid-tropospheric Aleutian low and the Western Pacific pattern from January to mid-March, leading to increased (decreased) planetary Wavenu-mber-1 activity in the stratosphere from mid- to late winter and thus more (less) winter stratospheric sudden warming events and late (early) onsets of SFW in early May (mid-April).

3.2 Influence of Lower Atmospheric Perturbation on Thermosphere/Ionosphere

Yu.[40]presented a multi-instrument experiment analysis of the intensification of metallic layered phenomena above thunderstorms. An enhanced ionospheric sporadic E layer with a downward tidal phase was observed followed by a subsequent intensification of neutral Na layer. The Na chemistry model simulation reproduced a consistent result of an enhanced Na layer by using the ionospheric observation as input. They found the enhancement of metallic layered phenomena above thunderstorms is associated with the atmospheric tides owing to the troposphere-mesosphere-ionosphere coupling.

Sun.[41]analyzed a special mid-latitudinal medium-scale traveling ionospheric disturbance (MSTID) event using OI (630.0 nm) all-sky airglow imager observations. The event was accompanied by a poleward surge of airglow depletion/enhancement and a bifurcation of depletion during the magnetically quiet period. The special structures were gene-rated at a decreasing height of ionosphere with the enhancement of nighttime plasma density and the increase of airglow intensity. They suggested that the poleward surge and bifurcation of airglow depletion were the result of the interaction between the passing-by MSTID and nighttime plasma density enhan-cement.

Based on four-year observations (2012–2015) from both all-sky airglow imager and the C/NOFS satellite, Wu.[42]studied the edge plasma enhan-cements of equatorial plasma depletions (EPEEPDs). The results showed that the EPEEPDs occurred only in the east and west edges of EPDs but not at the poleward edge. This was different from the earlier in-situ measurements obtained by low Earth orbit satellites. Both the all-sky airglow image observation and C/NOFS observation showed that the plasma enhancement was a high-incidence phenomenon, and the average incidence reached about 82% during days with EPDs observed. The zonal extension of EPEEPDs at different altitudes showed different scale characteristics. They suggested that the generation mec-hanism of EPEEPDs was possibly related to the polarized electric field of EPDs.

By combining observations from three meteor radars and an MST radar established by the Chinese Meridian Project with reanalysis data, Yu.[43]studied the SFW in 2015 spring and PW activities from the troposphere to the MLT at different latitudes. By means of the two warming events separated by only several days, the polar mean temperature increases by nearly 20 K at 10 hPa level, and the mean zonal wind decreases from more than 30 m·s–1to about –10 m·s–1, thus seasonal transition of the polar circulation is completed. The investigation shows that the Q10DW and Q16DW occur around the SFW. In the troposphere, their amplitudes are close to 10 m·s–1in the wind field. At 10 hPa level of the stratosphere, the strong wave activities arise before the SFW, while in the MLT, the waves are amplified following the SFW with the amplitude peaks about 10 days after the SFW onset. The wave amplitude in the MLT tends to increase in the zonal wind but decrease in the meridional wind with decreasing latitude, which is roughly consistent with the Hough modes. Hence, the Q10DW and Q16DW in the MLT are distinct from those in the stratosphere, and they are likely to be generated and strengthened in situ in the upper stratosphere and MLT.

Using measurements of meteor radar and MST radar established by the Chinese Meridian Project and reanalysis data, Huang.[44]report an ISO with about 30-day period at mid and high latitudes. Radar observations indicate that in the troposphere, the oscillation attains an amplitude peak in zonal wind at about 9 km, and propagates downward be-low 9 km. At about 9~16 km, the oscillation gradually decays with height, and then strengthens again as it propagates upward in the stratosphere. In the mesosphere, the oscillation obviously appears at 78~ 86 km with an maximum amplitude at 80 km. Rea-nalysis data show that in the troposphere, the oscillation propagates southward. At about 100 to 10 hPa levels, the oscillation is gradually reflected back to propagate northward, and then propagates poleward at higher altitude. Refractive index can explain its complex propagation characteristics very well. Consistency and coherence of its phase progression indicate that in the lower atmosphere, the oscillation comes from the polar region. Hence, ISOs can not only originate from but also propagate in the atmosphere at mid and high latitudes.

3.3 Gravity-wave Processes

Chen.[45]found that the time series of Gravity Wave Square Temperature Amplitude (GWSTA) and absolute Gravity Wave (GW) Momentum Flux (GWMF) at a certain altitude and latitude results from the complex interplay of GW sources, propagation through and filtering in lower altitudes, obli-que propagation superposing GWs from different source locations, and, finally, the modulation of the GW spectrum by the winds at a considered altitude and latitude. The strongest component is the annual variation, dominated in the summer hemisphere by subtropical convective sources and in the winter hemisphere by polar vortex dynamics. At heights of the wind reversal, a 180° phase shift also occurs, which is at different altitudes for GWSTA and GWMF. In the intermediate latitudes, a semiannual variation (SAV) is found. Dedicated GW modeling is used to investigate the nature of this SAV, which is a different phenomenon from the tropical SAO also seen in the data. In the tropics, a stratospheric and a mesospheric QBO are found, which are, as expected, in antiphase.

Hu.[46]showed that the stratospheric wave intensity (SWI) in December during 2001–2015 weakens, which is opposite to that during 1979–2000, implying a shift around the 2000s. The weakened SWI is dominated by its Wave Number-1 component, which is related to the wave propagation impeded by the decreased refractive index at high latitudes and the weakened wave activity in the troposphere. Changes in refractive index are mainly contributed by changes in the meridional potential vorticity gra-dientthe barotropic term. This shift in the wave number-1 waves leads to a shift in the stratospheric Arctic temperature, that is, warming (cooling) during 1979–2000 (2001–2015), implying that similar shifting phenomena may appear regardless of the continued greenhouse gas emissions.

Wu.[47]performed the Weather Research and Forecasting (WRF) simulations with five different spatial resolutions (25, 20, 15, 10, and 4 km) to examine the scale interactions between the GWs resolved by high-resolution Whole Atmosphere Com--munity Climate Model (WACCM) and those with smaller scales, its potential impact on the resolved waves, and the dependence of wave characteristics on spatial resolution. The sensitivity of GW structure to the choice of model horizontal resolutions is exa- mined, the simulation reveals more power at shorter horizontal wavelengths of GWs at finer resolutions. The magnitude of the zonal momentum flux calculated from the high-resolution WACCM is greater than the WRF-25 simulation, in addition, the magnitude is comparable to the results from the WRF-15 and WRF-10 simulations.

Relative to many investigations IGWs in the Antarctic, IGW activity in the Arctic region was paid less attention to. Huang.[48]use radiosonde observations at the Ny-Alesund station to investigate the IGW characteristics in the lower stratosphere over the Arctic. The observation reveals a prevailing eastward zonal wind below 20 km and an obvious annual cycle of the temperature from the troposphere to the lower stratosphere. By combining Lomb-Scargle spectrum and hodograph technique, the case study demonstrates that the lower stratospheric IGWs exhibit a feature of freely propagating waves. Stati- stical analysis indicates that the IGWs have a domi- nant horizontal (vertical) wavelength of 50~1050 km (1~4 km), and ratio (1~2.5) of the intrinsic to inertia frequencies. Wave energy exhibits an annual oscillation with the maximum in winter and the minimum in summer. In winter, the downward pro-pagating waves increase to about 20% due to polar stratospheric vortex. Because of the lower atmospheric filtering, the IGWs display a dominant direction of westward propagation, thus have a mean vertical flux of –0.647 mPa for the zonal momentum, which indicates that the IGWs can put a westward drag on the atmospheric wind field over the Arctic as they dissipate. All the vertical wavenumber spectra have spectral slopes from –2.23 to –2.99 close to the universal spectrum index of –3.

Gong.[49]studied a low-frequency inertial Atmospheric Gravity Wave (AGW) event with the data from three deferent detection tools, which are lidar, meteor radar, and TIMED/SABER. Observations from these three different instruments were compared, and it was found that signatures in the temperature perturbations and horizontal winds were induced by identical AGWs. According to these coor-dinated observation results, the horizontal wavelength and intrinsic phase speed were inferred to be about 560 km and about 21 m·s–1, respectively. Ana-lyses of the Brunt-V?is?l? frequency and potential energy illustrated that this persistent wave propagation had good static stability.

Based on the OH all-sky airglow imager observation over Kazan (55.8°N, 49.2°E), Russia, between August 2015 and July 2016, Li.[50]studied the mesospheric Gravity Waves (GWs) in the latitude band of 45°–75°N. The observed GWs showed a strong preference of propagation toward northeast in the whole year. This was significantly different from the results for other latitudes, where the propagation directions depended on season. It was reported that deep tropospheric convections were the dominant source of the GWs in spring, summer, and autumn. The results indicated that convection might also be an important source of GWs in the higher latitudes. Jet stream systems were considered as the generation mechanism of the GWs in winter.

Li.[51]studied a mesospheric bore event occurring on the night of 16–17 December 2014 by using the observations from both OH and OI (557.7 nm) all-sky airglow imagers in Lhasa (29.66°N, 90.98°E), the NPP and TIMED satellites, and a Doppler meteor radar. The results indicated that the winds in the height range of the OH layer were almost orthogonal to the propagation direction of the mesospheric bore. Both hydraulic jump theory and observations indicated that the duct initially shrank and followed by an expansion. The duct was strong in the OH layer but weak in the OI layer. The horizontal wavelengths and phase speeds of the bore packet decreased as the duct shrank and increased as the duct expanded. The intensity amplitude of the bore packet decreased slowly and then decreased sharply after dissipation. The bore might have leaked out of the duct with the variation of the depth of the duct.

Based on the OH all-sky airglow imager and the TIMED/SABER satellite observations and the ECMWF data, Wang.[52]studied the sources of two gravity wave events, which occurred on the night of 15–16 September 2013, using a method of reverse ray tracing. The results indicated that the wave source location of one event was highly related with intensity convective activity; the other might be related to dynamic instability due to vertical motion of air masses or convection activity.

Liu.[53]studied the seasonal and height dependencies of the orographic primary and larger- scale secondary Gravity Waves (GWs) by using the temperature profiles observed by the TIMED/SABER satellite from 2002 to 2017. The results indicated that at about 40°S and during Southern Hemisphere winter, there was a strong GW peak over the Andes Mountains which could extend to an altitude of about 55 km. The orographic GWs breaking occu-r-red above the peak height of the stratospheric jet. At the altitude of 55~65 km, the body forces produced by GW breaking and momentum deposition genera-ted larger-scale secondary GWs. At middle latitudes during summer, orographic GW breaking also gene-rated larger-scale secondary GWs which could pro-pagate to higher altitudes.

In addition, Lai.[54]developed a program to automatically extract gravity wave patterns from the all-sky airglow imager observation. The auto-extrac-tion program included a classification model based on a convolutional neural network and an object de-tection model based on a faster region-based convolu-tional neural network. In addition, the program can remove the interference of the wavelike mist near the imager.

4 Coupling between Stratosphere and Troposphere

Huang.[55]examined the preconditioning of events in which the Arctic stratospheric polar vortex shifts toward Eurasia, North America, and the Atlantic. They found that certain patterns of anomalous tropospheric blocking over northern Europe, the Bering Strait, and the eastern North Atlantic can be taken as the potential blocking precursors of the stratospheric polar vortex shifting toward Eurasia, North America, and the Atlantic.

Huang and Tian[56]analyzed the differences and similarities of Eurasian cold air outbreaks under the weak, strong, and neutral stratospheric polar vortex states. They found that the preexisting negative North Atlantic Oscillation pattern, the considerably negative stratospheric Arctic Oscillation signals entering the troposphere and positive sensible heat flux anomalies in the later stages of cold air outbreaks play important roles in the differences and similarities of these three types of cold air outbreaks.

Li.[57]investigated the connections between the first two Principle Components (PCs) of the SST anomalies over the North Pacific and the SSWs in the Northern Hemisphere winter. Their results showed that the SSW event occurs more frequently and is longer during the positive phases of PC2 than the negative phases of PC2. Moreover, they found that the positive phases of PC2 are marked by more positive Pacific–North America (PNA) and Western Pacific (WP) teleconnections in the upper troposphere.

Liang.[58]analyzed the upper tropospheric and stratospheric signals before the “June 5th” heavy rainfall event in South China in 2005. Their results showed that before the heavy rainfall there is a lower-lift-lower trend of tropopause over the rainfall region, stronger and more northward easterly winds in the UTLS over the subtropical region and the eastward shifted center of the South Asian High.

Luo.[59]investigated seasonal features of the tropopause fold events over TP and found that sha-llow tropopause folds occur mostly in spring while medium and deep folds occur mostly in winter. The relatively high-frequency areas of medium and shallow folds are located over the southern edge of TP. The region of high-frequency tropopause folds is loca-ted in the southern portion of the plateau in spring and moves northward in summer, controlled by the westerly jet movement.

Wang.[60]investigated the influence of the occurrence frequency of the MJO phases on the interannual variability of stratospheric wave activity in the mid-high latitudes of the Northern Hemisphere during boreal winter. The occurrence frequency of MJO Phase 4 in winter is significantly positively correlated with the interannual variability of the EP flux divergence anomalies in the northern extratro-pical stratosphere, while the occurrence frequency of MJO Phase 7 has an opposite and weaker effect on wave activities in the northern extratropical stratosphere.

Zhang.[61]found that the weakened/en-hanced Arctic polar vortex in the lower stratosphere during easterly/westerly QBO (EQBO/WQBO) phases is more noticeable in January and February than in November and December. The Arctic polar vortex shows a shift toward the Eurasian continent and away from North America in winter during EQBO phases compared with that during WQBO phases, with a greater shift in January and February than in November and December.

Zhang.[62]found that there is a dipole-like structure of geopotential height anomalies over the North Pacific Ocean during weak stratospheric polar vortex events. When the stratospheric polar westerly is decelerated, the high-latitude eastward waves slow down and equatorward propagation of eddy momentum flux at 60°N enhances. Hence the tropospheric eddy-driven jet over the North Pacific Ocean weakens and shows a southward displacement, lea-ding to the dipole in geopotential height via geos-trophic equilibrium.

Zhang.[63]found that the upward wave fluxes entering the stratosphere are stronger and more persistent during the downward-extending negative NAM events than during the non-downward-exten-ding negative NAM events. And the tropospheric wave intensity plays a more important role than the tropospheric conditions of planetary wave propagation in modulating the upward wave fluxes into the stratosphere.

Han.[64]found that the tropical tropopause temperature is uncorrelated with stratospheric water vapor in 1996. They showed that the instantaneous intensity of four short periods of deep convective activity, caused by strong surface cyclones and high sea surface temperatures can, on the one hand, transport water vapor in the troposphere directly into the lower stratosphere, and on the other hand, lift the tropopause and cool its temperature.

Li.[65]investigated the changing trend of ozone and its influence factors over Beijing during 2003–2013. Their results showed that the tropospheric ozone over Beijing increased significantly during 2003–2013. Further, they found that surface emission contributed to around 60% of the increase of tropospheric ozone over Beijing, while downward transmission from stratosphere and horizontal tran-sport contribute to about 20% and 10% of the increase of tropospheric ozone, respectively.

Luo.[66]found that the day-to-day changes in carbon monoxide (CO) and ozone (O3) tracer distributions in the UTLS are consistent with the dynamical variability during the Asian Summer Monsoon (ASM) season. The CO vertical cross sections from the Infrared Atmospheric Sounding Interferometer (IASI) combined with the daily maps provide the first observational evidence for a model analyses-based hypothesis on the preferred ASM vertical transport location and the subsequent hori-zontal redistributioneast-west eddy shedding.

Sang.[67]investigated how overshooting convection affects the water vapor content in the lower stratosphere. Their results showed that the net effect of overshooting convection on the lower stra-tospheric water content is moistening. Also they found that convective intensity is directly related to the effect of overshooting convection on the lower stratospheric humidity. They discovered that changes in vertical wind shear near the tropopause have no significant impact on the extent of overshooting but have important impacts on cross-tropopause water vapor exchange.

Han.[68]confirmed that the significant hemi-spheric asymmetries in the trends of N2O, CH4and HCl existed in the mid-latitude middle and lower stratosphere during the period of 2004–2012. They showed that changes in the vertical and meridional transport due to the residual circulation contribute to the observed asymmetric hemispheric trends of stratospheric trace gases. Southward shift of the upwelling branch of the residual circulation in recent decades partly explains the trends of these trace gases in the middle stratosphere while the eddy mixing has a small effect.

Wang.[69]investigated solar signals in the atmosphere and the ocean. Results indicate that the 11-year solar cycle is related to anomalous positive SST anomalies in the central Pacific which resemble an El Ni?o Modoki event. Such SST anomalies are amplified by a positive feedback through oceanic subsurface currents and heat transport in the equatorial Pacific and in turn modify the circulation and convection in the troposphere, resulting in lagged solar signals in tropical tropopause heights and tem-peratures as well as lower stratospheric water vapor.

Wang.[70]highlighted the importance of the vertical resolution in model development and climate change assessment. They found that a model with High Vertical-Resolution (HV-Res) gives a better representation of tropical Tropopause Temperature (TPT) in absolute values and seasonal variations. The corresponding changes in TPTs associated with SST anomalies are 30% stronger and more realistic in the HV-Res model.

Wang.[71]reported a severe surface ozone pollution episode over the YRD in 2018 spring associated with a deep SI event. This SI event is caused by a strong horizontal-trough, which brought ozone- rich air from stratosphere to troposphere. They de-monstrated that deep SIs contribute about 15 ppbv in spring to surface ozone variations in eastern China.

In recent years, several studies have documented the impact of stratospheric dynamic processes on tropospheric climate variability in the extra-tropics. Wei.[72]investigated the effect of a well-resolved stratosphere on East Asian winter climate by using the model outputs from the Coupled Model Inter- comparison Project phase 5 (CMIP5). A comparison between models with and without a well-resolved stratosphere revealed that the models with model top above the stratopause had a better simulation of the distribution of surface air temperature, sea level pressure, and precipitation than the models with a low-top below the stratopause. The difference of the East Asian winter climate between High-Top (HT) and Low-Top (LT) CMIP5 models is also evident in the future projection under higher (RCP85) and midrange (RCP45) emission scenarios. The HT mo-dels present about 1.3 and 1.7°C higher surface air temperature in East Asian region under RCP45 and RCP 85 scenarios by the end of this century than that of the LT models, respectively. As climate models have now become one of the primary tools in climate change assessment and projection, which was used as the base for mitigation measure and adaption policy. The results suggested that insufficient representation of the stratosphere might lead to unde-restimation of the anthropogenic global warming in a regional scale and hence had the potential to lead to insufficient response action and mitigation measures.

The Arctic Oscillation (AO) is the most dominant low-frequency mode of atmospheric variability in the extratropical Northern Hemisphere, which can strongly influence the global and regional climate through its teleconnections. Gong.[73]investigated the patterns and teleconnections of the winter mean AO based on observational and reanalysis datasets. They found that the Atlantic center of the AO pattern remains unchanged throughout the period of 1920–2010, whereas the Pacific center of the AO is strong during 1920–1959 and 1986–2010 and weak during 1960–1985. Gong.[74]further eva-luated the CMIP5 model performance in simulating the wintertime AO pattern. The magnitude of the North Pacific center of the AO pattern is shown to vary largely among the models, which is primarily modulated by the strength of the stratospheric polar vortex. A stronger stratospheric polar vortex can induce more planetary waves to reflect from the North Pacific to the North Atlantic and more wave activity fluxes to propagate from the North Pacific to the North Atlantic in the Strong Polar Vortex (SPV) models than in the Weak Polar Vortex (WPV) models. Hence, the coupling of atmospheric circulation between the North Pacific and North Atlantic is stronger in the SPV models, which may induce a stronger North Pacific center in the AO pattern. The increase in vertical resolution may improve the simulation of the stratospheric polar vortex, and thereby reduces the model biases in the North Pacific–North Atlantic coupling and the amplitude of the North Pacific center of the AO pattern in models.

Previous studies have found that the AO may also influence the atmosphere-ocean systems over the tropics. For instance, boreal spring AO may contribute to the occurrence of an El Ni?o event during the following winter through inducing westerly wind anomalies over the western tropical Pacific[75]. A recent study of Chen.[76]revealed a pronounced enhancement of the linkage between autumn Arctic sea ice concentration (ASIC) changes and following spring AO since the mid-1990s. After the mid-1990s, there exists a significant connection between interannual variation of the autumn ASIC and subsequent spring AO. During this period, autumn ASIC decrease results in pronounced tropospheric warming over the high latitudes and a reduction of the meridional temperature gradient. This leads to a decrease in circumpolar westerly winds and an increase in the upward propagation of quasi-stationary pla-netary waves, which weakens stratospheric polar vortex. Hence, a negative spring AO phase is generated via the downward propagation of the easterly wind anomalies. The results suggest that the interannual variability of autumn ASIC is much larger after the mid-1990s, which contributes to stronger stratospheric response and ASIC-spring AO connection.

The tropical convection anomaly, for example, associated with El Ni?o-Southern Oscillation (ENSO), plays an important role in the formation of extra-tropical teleconnections. Ding.[77]investigated the distinct patterns of boreal winter convection anomalies over the tropical Pacific and associated wave trains in the extratropics, and identified five major categories of tropical convection anomalies. Mechanisms for the formation of quasi-stationary extratropical wave trains associated with distinct winter patterns of seasonal mean tropical Pacific convection anomalies were further studied by utilizing observational and six Atmospheric Model Inter-comparison Project Phase 5 (AMIP5) high- skilled models datasets[78]. They found that the stratosphere-troposphere interaction plays an important role for extratropical atmospheric circulation anomalies. The stratospheric polar vortex not only modulates the underlying wave train, especially for the North Atlantic/Europe sector, but is also affected by the upper-tropospheric disturbance in high-latitudes through the upward propagating quasi-stationary planetary waves.

Huang.[79]demonstrated that the cross tropopause wind shear associated with stratospheric quasi-biennial oscillation (QBO) could impact the convection over the western North Pacific (WNP) and the resultant tropical cyclone (TC) genesis in May. The results indicate that the strong cross- tropopause shear over the key region (0°–5°N, 160°– 180°E) may suppress convection over the tropical WNP, especially over the South China Sea and the Philippines. This cross-tropopause wind shear is negatively correlated with TC genesis in May, with a decreased (increased) number of TCs corresponding to strong (weak) cross-tropopause wind shear. They suggested that this cross-tropopause wind shear can be treated as the combined impacts of the ENSO events and the QBO. The energy analysis indicates that the combined impacts of the decaying El Ni?o events and the QBO easterly phase might suppress the barotropic eddy kinetic energy conversion in May, whereas the decaying La Ni?a events and the QBO west phase act in an opposite manner.

Huang.[80]synthesized recent progress in researches on the atmospheric dynamics in the stra-tosphere and its dynamical interaction with the tro-pospheric processes. They focused on the dynamics of quasi-stationary planetary waves, the wave-basic flow interaction in the tropical stratosphere, the impact of atmospheric circulation variability in the stratosphere on circulation variability and climate in the troposphere, the numerical simu-lation of strato-spheric atmosphere and climate projection in the stratosphere under the background of global warming. The importance of variability of planetary wave activity and its association with the northern annular mode, a wider spectrum of gravity waves in forcing a realistic QBO and global meridional circulation, the cooling trend in the stratosphere and stratospheric processes for weather and climate anomalies near the surface was highlighted. More in-depth studies on the atmospheric dynamics in the stratosphere and improvements of the model performance in the strato-sphere were suggested in the future.

The Stratospheric Arctic Vortex (SAV) plays a critical role in forecasting cold winters in northern mid-latitudes. Its influence on the tropospheric mid- and high-latitudes has attracted growing attention in recent years. However, the trend in the SAV during the recent two decades is still unknown. Here, using three reanalysis datasets, Hu.[81]found that the SAV intensity during 1998–2016 has a strengthening trend, in contrast to the weakening trend before that period. Approximately 25% of this strengthening is contributed by the warming of Sea-Surface Temperature (SST) over the Central North Pacific (CNP). Observational analysis and model experiments show that the warmed CNP SST tends to weaken the Aleutian low, subsequently weakening the upward propagation of wavenumber-1 planetary wave flux, further strengthening the SAV. This strengthened SAV suggests important implications in understan-ding the Arctic warming amplification and in predicting the surface temperature changes over the northern continents.

Hu.[82]examined the decadal relationship between the Stratospheric Arctic Vortex (SAV) and Sea Surface Temperature Anomalies (SSTAs) in the North Atlantic and the dynamic mechanisms involved in the linkage between the two. Their results showed that there is a significant decadal linkage between SSTAs over the North Atlantic and the SAV, where warmed (cooled) SSTAs over the North Atlantic in association with its principal mode correspond to a weakened (strengthened) SAV. The warmed North Atlantic SSTAs tend to result in a weakened SAVtwo dynamic processes: (i) constructive interference at high latitudes with a ridge in the Atlantic sector and a trough in the Pacific accompanied by a negative North Atlantic Oscillation- like pattern over the North Atlantic and a weakened Aleutian low over the North Pacific; and (ii) more wavenumber-1 waves propagated into the Arctic stratosphere by modifying the baroclinic term of the zonal mean background state and altering the pro-pagating conditions around the tropopause over the Arctic. Results from reanalysis and model simulations both suggest that a strengthening wave intensity in the high-latitude troposphere and more upward propagation of the planetary wavenumber-1 wave in response to the warmed North Atlantic SSTAs conjunctly contribute to the increased planetary wave flux in the Arctic stratosphere, facilitating a weakened SAV.

Hu and Guan[83]reported a significant in-phase relationship between the PDO and stratospheric Arctic Vortex (SAV) on decadal time scales during 1950–2014, that is, the North Pacific Sea Surface Temperature (SST) cooling (warming) associated with the positive (negative) PDO phases is closely related to the strengthening (weakening) of the SAV. This decadal relationship between the North Pacific SST and SAV is different from their relationship on sub-decadal time scales. Observational and modeling results both demonstrated that the decadal variation in the SAV is strongly affected by the North Pacific SSTs related to the PDO via modifying the upward propagation of planetary Wavenumber-1 waves from the troposphere to the stratosphere. The decreased SSTs over the North Pacific tend to result in a deepened Aleutian low along with a strengthened jet stream over the North Pacific, which excites a weakened western Pacific pattern and a strengthened Pacific–North American pattern. These tropospheric circulation anomalies are in accordance with the decreased Refractive Index (RI) at middle and high latitudes in the northern stratosphere during the positive PDO phase. The increased RI at high latitudes in the upper troposphere impedes the planetary Wavenumber-1 wave from propagating into the stratosphere, and in turn strengthens the SAV. The responses of the RI to the PDO are mainly contributions of the changes in the meridional gradient of the zonal-mean potential vorticityalteration of the baroclinic term.

Using the CMIP5 Multimodel Ensemble (MME) historical experiments, the modulation of the stra-tospheric El Ni?o–Southern Oscillation (ENSO) teleconnection by the Pacific decadal oscillation (PDO) is investigated by Rao.[84]. El Ni?o (La Ni?a) significantly impacts the extratropical strato-sphere mainly during the positive (negative) PDO in the MME. Although the composite tropical ENSO SST intensities are similar during the positive and negative PDO in models, the Pacific–North America (PNA) responses are only significant when the PDO and ENSO are in phase. The local SST anomalies in the North Pacific can constructively (destructively) interfere with the tropical ENSO forcing to influence the extratropical eddy height anomalies when the PDO and ENSO are in (out of) phase. The difference between the positive and negative PDO in El Ni?o or La Ni?a winters filters out the tropical SST forcing, permitting the deduction of the extratropical SST contribution to the atmospheric response. The com-posite shows that the cold (warm) SST anomalies in the central North Pacific associated with the positive (negative) PDO have a similar impact to that of the warm (cold) SST anomalies in the tropical Pacific, exhibiting a positive (negative) PNA-like response, enhancing (weakening) the upward propagation of waves over the western coast of North America. The composite difference between the positive and negative PDO in El Ni?o or La Ni?a winters, as well as in eastern Pacific ENSO or central Pacific ENSO winters, presents a highly consistent atmospheric response pattern, which may imply a linear interference of the PDO’s impact with ENSO’s.

To show the typical spatiotemporal evolution of circulation anomalies during the NAM’s life cycle, Yu.[85]revisited the various stratosphere-tropo-sphere coupling relation from the perspective of the meridional mass circulation, and indicated that there is a large case-to-case difference in the temporal evolution and vertical profile of polar temperature anomalies during NAM events, which shows no strong dependence on the intensity and duration of NAM events, but agrees well with the variations of the three branches of mass circulation at 60°N: the stratospheric poleward warm air branch (ST), the poleward warm air branch in the upper troposphere (WB), and the equatorward cold air branch in the lower troposphere (CB). The various relationships among the three mass circulation branches are attributed to anomalous wave activities. The amplitude and westward tilt of waves are always stronger (weaker) throughout the stratosphere before (after) the peak time of negative NAM events, leading to a stronger (weaker) ST before (after) the peak time. Variations in WB and CB are mostly dependent on wave variabilities in the mid- to lower-troposphere, leading to variations in the timing of in or out-of- phase coupling of the ST with the WB and CB, and thus various thermal structures during NAM events.

[1] XIA Y, HUANG Y, HU Y,. Impacts of tropical tropopause warming on the stratospheric water vapor [J]., 2019, 53:3409-3418

[2] XIA Y, HU Y, LIU J,. Stratospheric ozone-induced cloud radiative effects on Antarctic sea ice [J]., 2020, 37:1-10

[3] LIU R, FU Y. Verification of an approximate thermodynamic equation with application to study on Arctic stratospheric temperature changes [J].., 2019, 76:3-9

[4] SHI C, GAO Y, CAI J,. Response of the dynamic and thermodynamic structure of the stratosphere to the solar cycle in the boreal winter [J].., 2018, 169:122-129

[5] HU D, GUAN Z, TIAN W. Signatures of the arctic stratospheric ozone in northern hadley circulation extent and subtropical precipitation [J].., 2019, 46:12340-12349

[6] MA J, SHANGGUAN M, XIA H,. Rayleigh and sodium lidar system incorporating time-division and wavelength- division multiplexing [J],, 2019, 448:116-123

[7] ZOU Z, XUE X, SHEN C,. Response of mesospheric HO2and O3to large Solar proton events [J].., 2018, 123:5738-5746

[8] XUN Y, YANG G, SHE C Y,. The first concurrent observations of thermospheric Na layers from two nearby central mid-latitude lidar stations [J]., 2019, 46(4):1892-1899

[9] JIAO J, YANG G, CHENG X,. Simultaneous lidar observation of peculiar sporadic K and Na layers at S?o José dos Campos (23.1° S, 45.9° W), Brazil [J]., 2018, 61(7):1942-1951

[10] WANG Z, YANG G, WANG J,. Lidar observations and studies of the lower-triangle potassium layer over Beijing [J]., 2018, 38(1):65-72

[11] YANG D, ZHANG T, WANG J, YAN C, PENG H. Observations of the prominent sporadic sodium layer over Haikou (in Chinese) [J]., 2018, 38(6):886- 890

[12] JIANG G, XU J, WANG W,. A comparison of quiet time thermospheric winds between FPI observations and model calculations [J]., 2018, 123. DOI.org/10.1029/2018JA025424

[13] LIU X, YUE J, WANG W,. Responses of lower thermospheric temperature to the 2013 St. Patrick’s Day geomagnetic storm [J]., 2018, 45:4656-4664

[14] LIU Y, XU J, LIU X,. Responses of multiday oscillations in the nighttime thermospheric temperature to solar and geomagnetic activities measured by Fabry-Perot interferometer in China [J]., 2019, 124. doi.org/10.1029/2019JA027237

[15] YANG R, XU J, ZHU Y, YUAN W. Comparison of Retrieval Methods for Neutral Wind Based on Airglow Measurements by a Ground-based Fabry-Perot Interferometer [J]., 2019, 39(1):76-83

[16], WEI K, WU X,. The effect of super volcanic eruptions on ozone depletion in a chemistry-climate model [J]., 2019, 36(8):823-836

[17] XIA Y, XU W, HU Y,. Southern-hemisphere high-latitude stratospheric warming revisit [J]., 2020, 54:1671-1682

[18] LI Xiaoting, TIAN Wenshou, XIE Fei,. Joint impacts of ENSO Modoki and QBO on stratospheric ozone in winter in the Northern Hemisphere [J].., 2019, 77(3): 456-474

[19] XIE Fei, LI Jianping, SUN Cheng,. Improved global surface temperature simulation using stratospheric ozone forcing with more accurate variability [J].., 2018, 8(1):1-10

[20] XIE Fei, MA Xuan, LI Jianping,. An advanced impact of Arctic stratospheric ozone changes on spring precipitation in China [J].., 2018b, 51(11/12):4029-4041

[21] XIE Fei, MA Xuan, LI Jianping,. Using observed signals from the arctic stratosphere and indian ocean to predict April–May precipitation in central China [J]., 2020, 33(1):131-143

[22] MA Xuan, XIE Fei, LI Jianping,. Effects of Arctic stratospheric ozone changes on spring precipitation in the northwestern United States [J].., 2019, 19(2):861-875

[23] XIE Fei, ZHOU Xin, LI Jianping,. Effect of the indo-pacific warm pool on lower-stratospheric water vapor and comparison with the effect of ENSO [J]., 2018, 31(3):929-943

[24] LU Jinpeng, XIE Fei, TIAN Wenshou,. Interannual variations in lower stratospheric ozone during the period 1984–2016 [J]., 2019, 124(14): 8225-8241

[25] ZHANG Jiankai, TIAN Wenshou, XIE Fei,. Stratospheric ozone loss over the Eurasian continent induced by the polar vortex shift [J].., 2018, 9(1):1-8

[26] ZHANG Jiankai, TIAN Wenshou, XIE Fei,. Zonally asymmetric trends of winter total column ozone in the nor-thern middle latitudes [J].., 2019, 52(7/8):4483-4500

[27] SHANGGUAN Ming, WANG Wuke, JIN Shuanggen. Variability of temperature and ozone in the upper troposphere and lower stratosphere from multi-satellite observations and reanalysis data [J].., 2019, 19(10):6659-6679

[28] XIAO Na, ZHANG Jiankai, TIAN Wenshou,. Effects of nitrogen oxide emissions over East Asia on ozone and temperature in UTLS region of the Northern Hemisphere [J]., 39(3):1-14

[29] YI W, XUE X, REID I,. Estimation of mesospheric densities at low latitudes using the Kunming meteor radar together with SABER temperatures [J]..:., 123: 3183-3195

[30] WU J, FENG W, XUE X,. The 27-day solar rotational cycle response in the mesospheric metal layers at low latitudes [J].., 2019, 46:7199-7206

[31] YI W, XUE X, CHEN J,. Quasi-90-day oscillation observed in the MLT region at low latitudes from the Kunming meteor radar and SABER [J]., 3:136-146

[32] YI W, XUE X, REID I M,. Climatology of the mesopause relative density using a global distribution of meteor radars [J].., 19:7567-7581

[33] RAO J, REN R, CHEN H,. Predictability of stratospheric sudden warmings in Beijing climate center forecast system with statistical error corrections [J]., 2019, 124:8385-8400

[34] RAO J, REN R, CHEN H,. Sub-seasonal to seasonal hindcasts of stratospheric sudden warming by BCC–CSM1.1 (m): a comparison with ECMWF [J]., 2019, 36(5):479-494

[35] YU Y, CAI M, SHI C,. Sub-seasonal prediction skill for the stratospheric meridional mass circulation variability in CFSv2 [J].., 2019, 52:631-650

[36] HU J, REN R. Stratospheric control of the Indian summer monsoon onset [J].., 2018, 83:135-147

[37] YU Y, CAI M, SHI C. On the linkage among strong stratospheric mass circulation, stratospheric sudden warming, and cold weather events [J]., 2018, 146: 2717-2739

[38] HU J, LI T, XU H. Relationship between the North Pacific Gyre Oscillation and the onset of stratospheric final warming in the northern hemisphere [J].., 2018, 51:3061- 3075

[39] RAO J, REN R, CHEN H,. The stratospheric sudden warming event in February 2018 and its prediction by a climate system model [J]., 2018, 123:13332-13345

[40] YU B, XUE X, KUO C,. The intensification of metallic layered phenomena above thunderstorms through the modu-lation of atmospheric tides [J].., 2019, 9(1):1-13

[41] SUN L, XU J, XIONG C,. Midlatitudinal special airglow structures generated by the interaction between propagating medium-scale traveling ionospheric disturbance and nighttime plasma density enhancement at magnetically quiet time [J]., 2019, 46. doi.org/ 10.1029/2018GL080926

[42] Wu K, Xu J, Xiong C, Yuan W. Edge plasma enhancements of equatorial plasma depletions observed by all-sky imager and the C/NOFS satellite [J]., 2018, 123:8835-8849

[43] YU F R, HUANG K M, ZHANG S D,. Quasi 10- and 16-day wave activities observed through meteor radar and MST radar during stratospheric final warming in 2015 spring [J]., 2019, 124. DOI.org/10.1029/ 2019JD030630

[44] HUANG K M, XI Y, WANG R,. Signature of a quasi 30-day oscillation at midlatitude based on wind observations from MST radar and meteor radar [J].,2019, 124. doi.org/10.1029/2019JD031170

[45] CHEN D, STRUBE C, ERN M,. Global analysis for periodic variations in gravity wave squared amplitudes and momentum fluxes in the middle atmosphere [J].., 2019, 37:487-506

[46] HU D, GUO Y, GUAN Z. Recent weakening in the stratospheric planetary wave intensity in early winter [J].., 2019, 46:3953-3962

[47] WU J F, XUE X H, LIU H L,. Assessment of the simulation of gravity waves generation by a tropical cyclone in the high-resolution WACCM and the WRF [J].., 2018, 10:2214-2227

[48] HUANG K M, YANG Z X, WANG R,A statistical study of inertia gravity waves in the lower stratosphere over the Arctic region based on radiosonde observations [J]., 2018, 123. doi.org/10.1029/ 2017JD027998

[49] GONG S, YANG G, XU J,. Gravity wave propagation from the stratosphere into the mesosphere studied with lidar, meteor radar, and TIMED/SABER [J]., 2019, 10(2):81

[50] LI Q, YUSUPOV K, AKCHURIN A,. First OH airglow observation of mesospheric gravity waves over European Russia region [J]., 2018, 123:2168-2180

[51] LI Q, XU J, YUE J,. Evolution of a mesospheric bore in a duct observed by ground-based double-layer imagers and satellite observations over the Tibetan Plateau region [J]., 2019, 124: 1377-1388

[52] WANG C M, LI Q Z, XU J Y,. A study of wave sources of gravity wave events observer by OH airglow imager located at Donggang station [J].., 2018, 61(6):2198-2206

[53] LIU X, XU J, YUE J,. Orographic primary and secondary gravity waves in the middle atmosphere from 16-year SABER observations [J]., 2019, 46:4512-4522

[54] LAI C, XU J, YUE J,. Automatic extraction of gravity waves from all-sky airglow image based on machine learning [J].., 2019, 11:1516

[55] HUANG Jinlong, TIAN Wenshou, GRAY L J,. Preconditioning of Arctic stratospheric polar vortex shift events [J]., 2018, 31(14):5417-5436

[56] HUANG Jinlong, TIAN Wenshou. Eurasian cold air outbreaks under different arctic stratospheric polar vortex strengths [J].., 2019, 76(5):1245-1264

[57] LI Yuanpu, TIAN Wenshou, XIE Fei,. The connection between the second leading mode of the winter North Pacific sea surface temperature anomalies and stratospheric sudden warming events [J].., 2018, 51(1/2):581-595

[58] LIANG Jinglin, LUO Jiali, TIAN Hongying,. Analysis of abnormal signals in the upper troposphere and stratosphere before the persistent heavy rainfall event in South China in June 2005 [J].., 24(2):237-250

[59] LUO Jiali, LIANG Wenjun, XU Pingping,. Seasonal features and a case study of tropopause folds over the Tibetan Plateau [J].., 2019. DOI: 10.1155/2019/4375123

[60] WANG Feiyang, TIAN Wenshou, XIE Fei,. Effect of Madden–Julian oscillation occurrence frequency on the interannual variability of Northern Hemisphere stratospheric wave activity in winter [J]., 2018, 31(13):5031-5049

[61] ZHANG Jiankai, XIE Fei, MA Zhichao,. Seasonal evolution of the quasi-biennial oscillation impact on the Northern Hemisphere polar vortex in winter [J]., 2019, 124(23):12568-12586

[62] ZHANG Kequan, WANG Tao, XU Mian,. Influence of wintertime polar vortex variation on the climate over the North Pacific during late winter and spring [J]., 2019, 10(11):670

[63] ZHANG Ruhua, TIAN Wenshou, ZHANG Jiankai,. The corresponding tropospheric environments during downward-extending and nondownward-extending events of strato-spheric northern annular mode anomalies [J]., 2019, 32(6):1857-1873

[64] HAN Yuanyuan, TIAN Wenshou, ZHANG Jiankai,. A case study of the uncorrelated relationship between tropical tropopause temperature anomalies and stratospheric water vapor anomalies [J]., 2018, 24(3):356-368

[65] LI Yang, ZHANG Jiankai, TIAN Wenshou,. Impact of dynamical transmission and surface emission on ozone change in troposphere over Beijing [J]., 2018, 36(2):157-166

[66] LUO Jiali, PAN L L, HONOMICHL S B,. Space-time variability in UTLS chemical distribution in the Asian summer monsoon viewed by limb and nadir satellite sensors [J].., 2018, 18(16):12511-12530

[67] SANG Wenjun, HUANG Qian, TIAN Wenshou,. A large eddy model study on the effect of overshooting convection on lower stratospheric water vapor [J]., 2018, 123(18):10023-10036

[68] HAN Yuanyuan, TIAN Wenshou, CHIPPERFIELD M P,. Attribution of the hemispheric asymmetries in trends of stratospheric trace gases inferred from Microwave Limb Sounder (MLS) Measurements [J]., 2019, 124(12):6283-6293

[69] WANG Wuke, SHANGGUAN Ming, TIAN Wenshou,. Large uncertainties in estimation of tropical tropopause temperature variabilities due to model vertical resolution [J].., 2019, 46(16):10043-10052

[70] WANG Wuke, MATTHES K, TIAN Wenshou,. Solar impacts on decadal variability of tropopause temperature and Lower Stratospheric (LS) water vapour: a mechanism through ocean-atmosphere coupling [J].., 2019, 52(9/10):5585-5604

[71] WANG Yiping, WANG Hongyue, WANG Wuke. A stratospheric intrusion-influenced ozone pollution episode associated with an intense horizontal-trough event [J]., 2020, 11(2):164

[72] WEI K, CAI Z, CHEN W, XU L. The effect of a well-resolved stratosphere on East Asian winter climate [J]., 2018, 51:4015-4028

[73] GONG H, WANG L, CHEN W, NATH D. Multidecadal fluctuation of the wintertime Arctic Oscillation pattern and its implication [J]., 2018, 31:5595-5608

[74] GONG H, WANG L, CHEN W,. Diversity of the wintertime Arctic oscillation pattern among CMIP5 models: role of stratospheric polar vortex [J]., 2019, 32:5235-5250

[75] CHEN S, YU B, CHEN W, WU R. A review of atmosphere-ocean forcings outside the tropical Pacific on the El Ni?o-Southern Oscillation occurrence [J]., 2019, 9:439

[76] CHEN S, WU R, CHEN W. Enhanced impact of Arctic sea ice change during boreal autumn on the following spring Arctic oscillation since the mid-1990s [J]., 2019, 53:5607-5621

[77] DING S, CHEN W, GRAF H F,. Distinct winter patterns of tropical Pacific convection anomaly and the associated extratropical wave trains in the Northern Hemisphere [J]., 2018, 51:2003-2022

[78] DING S, CHEN W, GRAF H F,. Quasi-stationary extratropical wave trains associated with distinct tropical Pacific seasonal mean convection patterns: observational and AMIP model results [J]., 2019, 53:2451-2476

[79] HUANGFU J, CHEN W, JIAN M, HUANG R H. Impact of the cross-tropopause wind shear on tropical cyclone genesis over the Western North Pacific in May [J]., 2019, 52:3845-3855

[80] HUANG R H, CHEN W, WEI K,. Atmospheric dynamics in the stratosphere and its interaction with tropospheric processes: progress and problems [J]., 2018, 42(3):463-487

[81] HU D, GUAN Z, TIAN W,. Recent strengthening of the stratospheric Arctic vortex response to warming in the central North Pacific [J]..,2018, 9:1697

[82] HU D, GUAN Z, GUO Y,. Dynamical connection between the stratospheric Arctic vortex and sea surface temperatures in the North Atlantic [J].., 2019, 32:6979-6993

[83] HU D, GUAN Z. Decadal relationship between the stratospheric Arctic vortex and Pacific Decadal oscillation [J]., 2018, 31:3371-3386

[84] RAO J, GARFINKEL C, REN R. Modulation of the northern winter stratospheric El Ni?o–Southern Oscillation teleconnection by the PDO [J]., 2019, 32:5761- 5783

[85] YU Y, REN R. Understanding the variation of stratosphere-troposphere coupling during stratospheric northern annular mode events from a mass circulation perspective [J].., 2019, 53:5141-5164

P 351

CHEN Zeyu, CHEN Hongbin, XU Jiyao, HUANG Kaiming, XUE Xianghui, HU Dingzhu, CHEN Wen, YANG Guotao, TIAN Wenshou, HU Yongyun, XIA Yan. Advances in the Researches of the Middle and Upper Atmosphere in China., 2020, 40(5): 856-874. DOI:10.11728/cjss2020.05.856

June 23, 2020

E-mail: chb@mail.iap.ac.cn