Back to Program


Search By:

Presentation Mode : All
Conference Day : 02/08/2021
Time Slot : AM2 11:00 - 13:00
Sections : AS - Atmospheric Sciences










Atmospheric Sciences | Mon-02 Aug




AS22-A003 | Invited
Multiscale Modeling for Urban Climate Study

Hiroyuki KUSAKA#+
University of Tsukuba, Japan


Meso-scale and micro-scale models have been largely contributing to urban meteorology and climatology. This talk will introduce highlights of multi-scale modeling studies by our group including urban canopy scheme (UCM), regional climate model (WRF), and city-scale Large Eddy Simulation (LES) model. The talk is not restricted to showing the development of the numerical models. Research results of urbanization impacts on low-level cumulus clouds and heavy precipitation will be discussed here. Additionally, high light results of future urban climate projection will be also introduced. Last, we will show you recent results of building-resolving simulations toward understanding better climate change adaptation strategies.

AS22-A010
Estimating Urban Surface Roughness Aerodynamic Parameters Using Random Forest

Guangdong DUAN1#+, Tetsuya TAKEMI2
1Dalian Maritime University, China, 2Kyoto University, Japan


The surface roughness aerodynamic parameters, z0 (roughness length) and d (zero-plane displacement height), are vital to the accuracy of the Monin-Obukhov similarity theory (MOST). Deriving improved urban canopy parameterisation (UCP) schemes within the conventional framework remains mathematically challenging. The current study explores the potentiality of a machine learning (ML) algorithm, a random forest (RF), as a complement to the traditional UCP schemes. Using large-eddy simulation (LES) and ensemble sampling, in combination with nonlinear least-squares regression of the logarithmic-layer wind profiles, a dataset of approximately 4500 samples is established for the aerodynamic parameters and the morphometric statistics, enabling the training of the ML model. While the prediction for d is not as good as the UCP scheme after Kanda et al. [Bound.-Layer Meteor. 148 (2013), pp. 357-377], the performance for z0 appears promising. The RF algorithm also categorises z0 and d with an exceptional performance score: the overall bell-shaped distributions are well predicted and the ±0.5σ category (i.e. the 38% percentile) competently captured (37.8% for z0 and 36.5% for d). Amongst the morphometric features, the mean and maximum building heights (Have and Hmax) are found to be of predominant influence on the prediction of z0 and d. A perhaps counter-intuitive result is the considerably less striking importance of the building-height variability, σH. Possible reasons are discussed. The feature importance scores could be of usefulness for identifying the contributing factors to the surface aerodynamic characteristics. Issues on overfitting are discussed. The results may shed some light on the development of ML-based UCP for mesoscale modelling.

AS22-A013 | Invited
Multiscale Modeling of Transboundary Transport of Air Pollution in 2015 Southeast Asia Biomass Burning

Xianxiang LI1#+, Chao YUAN2
1Sun Yat-Sen University, China, 2National University of Singapore, Singapore


Biomass burning is a major source of air pollution, especially at areas where heavy anthropogenic pollution is absent. Apart from its impact on the atmospheric radiative budget and air quality, biomass burning could have significant impact on human health at the downwind areas. High aerosol concentrations from burnings were reported to be responsible for persistent human exposure to unhealthy air quality environment. Biomass burning in Southeast Asia is featured with the periodic burnings in springtime in Indo-China (e.g. Vietnam, Thailand and Myanmar) and catastrophic burning events in Indonesia.  During September-October 2015, a severe biomass burning episode occurred in Southeast Asia, which impacted the air quality in countries in this region. This study employed a multiscale modeling approach to investigate the transboundary transport of air pollutant from this biomass burning and its impact to Singapore. At mesoscale, WRF-Chem model was used to perform a high-resolution (5 km) simulation of the meteorology and chemical processes during these two months. Two simulations were carried out: one with fire emissions from FINN (Fire Inventory from NCAR) (WRF-Fire), and the other without fire emission (WRF-NoFire). The results from WRF-Fire were validated against various observations, including in-situ and satellite measurements. The WRF-Chem results were then downscaled to feed into a computational fluid dynamics (CFD) model covering Singapore. Several parametric CFD studies were performed to investigate the effects of urban morphology, such as urban porosity and building height variance, on the air quality during this biomass burning episode. It was found that urban development, or high density planning at Singapore did not make the air quality worse during this period, since transboundary pollutant is at larger scale and well mixed by turbulence.  

AS22-A007
Large-eddy Simulation of Urban Boundary Layer Flow Over Dense Built City

Lan YAO+, Chun-Ho LIU#
The University of Hong Kong, Hong Kong SAR


Urban-boundary-layer (UBL) flows over real urban surfaces, the Mong Kok neighborhood in Kowloon Peninsula, Hong Kong, was investigated by large-eddy simulation (LES). Full-scale building models (average building height Have = 30m), covering from Tsim Sha Tsui to Sham Shui Po, were constructed. Prevailing inflow of Southerly winds (U = 10 m sec-1) was prescribed in neutral stratification. The turbulence statistics extracted from three (spatial) subdomains, each has the size X × Y × Z = 800 m × 100 m × 2000 m, were analyzed after validation against reduced-scale wind tunnel experiments (1:1,200). Inertial sublayers (ISLs) were identified within the range of 2.7Have ≤ z ≤ 4.0Have by fitting logarithmic law-of-the-wall (log-law). Thereafter, roughness sublayers (RSLs) resided in 3.3Have ≤z ≤5.3Have. The RSL flows were found faster and more uniform than those extrapolated by the log-law. A strong, elevated shear layer was developed (2.7Have ≤ z ≤ 3.3Have), featuring higher zero-plane displacement d and inflection of velocity <u >/U profile than those over idealized roughness elements. The third- and fourth-order moments of velocity statistics generalized rear, fast RSL entrainment and bulky, slow ISL detrainment over real urban surfaces. Conditional sampling further signified the importance of large-scale RSL sweep (u’’ > 0, w’’ < 0) and ISL ejection (u’’ < 0, w’’ > 0) structures in downward momentum transport. Minor discrepancies in ISL structure were observed due to a high-rise building (8.5Have) to the West. Twin peaks of turbulence intensity profiles <u’’u’’>1/2/U and -<u’’w’’>/U2 arose. Coherent structure of ejection was suppressed in the wake of the high-rise building. Besides, the (component of) spanwise momentum flux -<u’’v’’> was comparable to its (main) downward counterpart -<u’’w’’>. These distinctions should be considered for the analysis of UBL over real urban morphology.

AS22-A001 | Invited
Observation and Real-Time WRF-LES of Complex Microscale Flows at the Hong Kong International Airport

K.K. HON#+, P.W. CHAN
Hong Kong Observatory, Hong Kong SAR


The concurrent availability of advanced remote sensing instrument in the form of long-range Doppler LIDARs together with operational numerical weather prediction (NWP) system at sub-kilometer grid spacing at the Hong Kong International Airport (HKIA) has provided a unique opportunity in the study and validation of meso- and micro-scale airflow.  HKIA is constructed on an artificial island in close proximity to complex oropgraphy, and is well known for its susceptibility to airflow disturbances, due to terrain influence or other forcings, which are capable of affecting the landing and departing aircraft.  In this presentation, various airflow features around HKIA as revealed by Doppler LIDAR observations will be reviewed, together with an evaluation of the short-term forecasts of a 200-m WRF-LES suite in reproducing such features in a real-time or near real-time manner.  Limitations and way forward for real-time WRF-LES applications are also discussed.

AS22-A005
Parameterising Urban Flows with a Multiscale Eddy Viscosity

Wanting LIU+, Ghar Ek LAU, Keith NGAN#
City University of Hong Kong, Hong Kong SAR


Although flow over urban areas introduces complicated spatial interactions, urban canopy parameterisations typically rely on bulk or mixing-length models. Following the large-eddy simulation approach, however, parameterisations that represent interactions between resolved and subgrid scales are desirable. In this work, an effective eddy viscosity that represents interactions between the nominal mesoscale flow and subgrid urban scales is calculated for a large-eddy simulation of flow over a regular obstacle array. The use of orthonormal wavelets allows for the transfer or eddy viscosity to be analysed as a function of scale and position. The eddy viscosity is maximized around the roof level and decays approximately linearly with height in the inertial sublayer for a wind perpendicular to the obstacles; the extension to other wind directions is straightforward away from the roof level. Applications to mesoscale modelling with WRF are discussed.

AS17-A005
A Unique and Simple Technique to Detect the Chance of Significant Interactions Between Typhoons

Ravi Shankar PANDEY#+, Yuei-An LIOU
National Central University, Taiwan


The cyclone to cyclone interaction also called the Fujiwhara effect may cause a change in direction and strength of interacting typhoons, making them not only difficult to forecast, but also cause an unexpected level of disaster. This research provides a unique but simple technique of detecting the chance of significant interaction (partial or complete merger) between typhoons by the combination of three factors namely differential satellite images, sudden fluctuation in interacting typhoons’ parameters, and values of the Liou-Liu empirical formula. First, in October 2009, the Fujiwhara effect between typhoons Parma and Melor caused the former to stagnate around Northern Philippines for nearly five days and make three landfalls. Differential image processing techniques and masking are applied on RGB (Red, Green, Blue) composites to identify various kinds of interactions. The Liou-Liu empirical formula identified the significant Fujiwhara effect at 0600 UTC on 06 October 2009, while a sudden drop (incline) in maximum wind speed and incline (drop) in central pressure during significant interaction is observed in the typhoon Melor (Parma). Similarly, the differential images identify it as the partial merger kind of significant interaction at 0600 UTC on 06 October 2009. Second, hurricanes Laura and Marco (August 2020) in the Gulf of Mexico were tested using the same technique, where a chance of significant interaction was detected although, due to the early landfall of Marco and less warm ocean space available in the Gulf of Mexico, the interaction becomes weak. A unique feature of interacting unequal typhoons is observed by the above cases while significant interaction, the weaker typhoon\hurricane is partially strengthened by the comparatively stronger typhoons\hurricane. The cases of such interaction may be conveniently investigated by the abovementioned technique that combined the Liou-Liu formula, differential images, and checking of sudden variations in typhoons’ parameters in the future.

AS09-A006
Dust Emission and Transport in Northwest China: WRF-chem Simulation and Comparisons with Multi-sensor Observations

Jianqi ZHAO+, Xiaoyan MA#
Nanjing University of Information Science & Technology, China


In this study, the WRF-Chem model is employed to simulate a dust process in Northwest China during May 2018. The model's ability to simulate the dust process is firstly evaluated using various satellite-retrieved and observational data. The four-dimensional assimilation method is also used to optimize meteorological data and effectively improve the simulation of the dust process. The comparisons between the simulations based on five dust emission schemes within WRF-Chem and the observations show that, the Shao01 scheme overall has good performance in simulating the emission flux, the spatial pattern of source region and the spatiotemporal variation of dust mass concentration during this dust process. In comparison to Shao01, the GOCART AFWA and Shao04 schemes can also produce quite similar spatial pattern of dust source region, but tend to overestimate or underestimate dust emission and mass concentration. The Shao11 scheme fails to simulate the dust process since the importance of the fully disturbed particle size distribution is omitted. It is also noted that the GOCART scheme can well reproduce dust emission processes under weak wind erosion but underestimate dust emission flux under strong wind erosion. Northwest China is covered by mountains, basins, deserts and other landforms, thus the complex terrain is one of the key factors to influence the dust process over the region. Our study shows that after being emitted, the airborne dust transported toward the east and west. The dust to the east was diffused rapidly, but the portion toward the west was blocked and accumulated at the edges of the mountains and thus produced dust weather characterized by high dust concentration and long lifetime. The dust accumulated at the edges of the mountains could reach an altitude of more than 6 km due to wind and thermal effect, and finally arrive at Tibetan Plateau and eastern China.

AS09-A015
The Implementation of Yonsei Aerosol Retrieval (YAER) Algorithm to Gk-2a/ami and Fy-4a/agri.

1, 1, Hyunkwang LIM2, 1, P.W. CHAN3
1, , 2Yonsei University, Korea, South, 3Hong Kong Observatory, Hong Kong SAR


The Yonsei AErosol Retrieval Algorithm (YAER) has been improved with geostationary satellites such as GOCI/COMS (Lee et al., 2010; Choi et al., 2017) and AHI/Himawari-8 (Lim et al., 2018). Meanwhile, Korea Meteorological Administration (KMA) launched a new generation of the geostationary meteorological satellites GEO-KOMPSAT 2A (GK-2A) on December 5, 2018. China Meteorological Administration (CMA) launched Feng-Yun 4A (FY-4A) on 11 December 2016. In this study, we implement the YAER algorithm to both AMI and AGRI. Estimating accurate surface reflectivity is important when retrieving Aerosol optical depth (AOD) from a satellite. Surface reflectance estimation over land adopts a minimum reflectance method (MRM method). The ocean surface reflectance makes use of the Cox and Munk method. Current YAER algorithm can retrieve aerosol properties only over dark surfaces. Therefore, masking of bright surfaces including snow, desert, turbid water, and that of cloud pixels is important. With the 12 (AMI) and 11 (AGRI) infrared channels, the observed data from both sensors can mask those pixels with decent accuracy. Especially, detection of cirrus cloud pixels is more advantageous on the AMI and AGRI than the other satellites (e.g., GOCI, AHI) since they have a 1370nm shortwave infrared channel. Despite there are only 2 visible channels on the AGRI, while the AHI has 3, the retrieved AOD products still have a good qualitative agreement with the AHI. This study presents a preliminary result of the aerosol optical retrieval from AMI and AGRI. The retrieval of aerosol properties of the AMI and AGRI YAER algorithm will bridge the spatial and temporal gap of the GOCI and the AHI.

AS09-A036
Aerosol Optical Properties and Radiative Effects During Pollution Episodes in Beijing

Yuanxin LIANG1+, 2, Hong WANG1
1Chinese Academy of Meteorological Sciences, China, 2,


This research comprehensively using ground-based observations, remote sensing and model simulations to better reveal the important role of aerosol shortwave radiative effects in the pollution process. Based on continuous observations of aerosol optical properties from sun-photometer, the variation of aerosol optical depth, Angstrom exponents, single scattering albedo(SSA) and asymmetry factor during different pollution episodes in Beijing in 2018 are analyzed. Combined with Raman-Mie Lidar detection, the vertical variation of aerosol extinction coefficient is analyzed in detail. Using the observational data, the shortwave radiative heating rate(HR) during the pollution process are calculated by a shortwave radiative transfer model.
The main result shows that air pollution in Beijing which has the highest particulate concentration in the atmosphere is affected by mixing of coarse and fine particles. During the pollution process, aerosol concentration increases explosively and their extinction effect increases in troposphere, which reduces the solar radiation reaching the surface. However, the optical properties of different aerosol components are quite different, the absorption of dust aerosols causes SSA440nm decreased to 0.88, and the strong scattering of inorganic aerosols during haze episode caused SSA440nm above 0.94. Aerosol layer can heat the atmosphere evidently, the magnitude of radiative heating rates by aerosol depends on distribution of the aerosol in the vertical direction, while the heating rate under the concentrated heating zone decreases rapidly.

AS09-A038
The Deweathered Concentrations of NO2 and Its Trend at Old and Newly Developed Urban Center and Rural Background Sites in Qingdao

Qinchu FAN#+, Xiaohong YAO
Ocean University of China, China


In order to improve air quality in Qingdao, the local government has taken a number of measures to cut air pollutants’ emission in the last several years. However, it is challenging to quantify their effects because of climate anomaly. To decouple the influence of meteorological factors on concentrations of air pollutants and their long-term trends related to mitigation measures, here, we applied machine learning methods, i.e., random forests algorithm and boosted regression trees algorithm, to extract the deweathered trends in concentrations of air pollutants. Using NO2 as an example, we selected the NO2 data from 2014 to 2020 at three local monitoring sites such as Shinanxi, Chengyangbei and Yangkou, which represented the old downtown, the newly developed urban center and rural background atmospheric environments. After minimizing the impact of meteorological conditions, the effects of two anthropogenic activities can be clearly identified. One is a large decrease in NO2 concentration from June to October 2018 at the three sites during the Shanghai Cooperation Organization Qingdao Summit. The other was the sharp decline of NO2 concentration from January 2020 to the mid of 2020. Two method analysis results also reveal that the concentration of NO2 at Chengyang and Shinanxi showed a slightly downward trend from 2014 to 2020, but a steady trend was identified at Yangkou.  In addition, we compared the analysis results of the two methods and found that the random forests method has a better performance to eliminate those abnormal perturbations.

AS09-A045
Analysis on Combustion Characteristics Over East Asia Using Satellite Datasets

Jaemin HONG1+, 2, 2
1Yonsei University, Korea, South, 2,


In recent decades, East Asia has experienced significant air pollution problems despite the efforts to reduce emission of pollutants. For an effective regulation of emission, the characteristics of various combustion processes are needed to be understood as they are accompanied by general anthropogenic activities. Those characteristics can be represented by the emission factor, and can be inferred from the ratio of enhancement in each species in the atmosphere (i.e. NO2 to CO2, CO to CO2), which can be observed from space. In this research, multiple satellite datasets were used to estimate enhancement of various species in East Asia, which includes carbon dioxide from the Orbiting Carbon Observatory-2 (OCO-2), nitrogen dioxide from the Ozone Monitoring Instrument (OMI), and carbon monoxide from the Measurement Of Pollution In The Troposphere (MOPITT). The enhancements of these species were calculated as the difference of abundance in urban and rural areas. To avoid noise in the estimation of enhancements and their ratios, we used long-term datasets and filtered data by wind speed. We found different ratios in each cities, indicating different characteristics of combustion.  

AS09-A048
Trend Analysis of Multiple Air Pollutants Over East Asia Based on Satellite Observations

Dha Hyun AHN1#+, Heesung CHONG1, 2, 2, 2
1Yonsei University, Korea, South, 2,


The economic development of East Asia had resulted in the increased emission of air pollutants, and each country has regulated the pollution emissions by implementing air quality improvement policies. In this study, the trends of various air pollutants across East Asia were analyzed and the characteristics of variation in each city were described. To examine the improvement of air quality after the implementation of the air pollution regulation, satellite observation data were used for the last 8 years (MAR 2011-FEB 2019). Aerosol optical depth (AOD) from the Geostationary Ocean Color Imager (GOCI), ammonia (NH3) from the Infrared Atmospheric Sounding Interferometer (IASI), carbon monoxide (CO) from the Measurements Of Pollution In The Troposphere (MOPITT), sulfur dioxide (SO2), tropospheric nitrogen dioxide (NO2), tropospheric ozone (O3), and formaldehyde (HCHO) from the Ozone Monitoring Instrument (OMI) were used in this study. Since the concentrations of aerosol, ozone and trace gases are dependent on the meteorological factors, the impact of meteorological factors was removed using the multiple linear regression method. The meteorological factors were obtained from ERA5 reanalysis data developed in European Center for Medium-Range Weather Forecasts (ECMWF). As a result, it was found that all components except NH3 showed a decreasing trend over the past 8 years. In particular, the trends of increasing NH3 and decreasing the other pollutants were strongest in northeastern China, and relatively weak in the Korean Peninsula and Japan. The trends in the pixels over the ocean followed the trend that appeared in China, and the magnitude of trends were maintained across East Asia especially for the substances having longer lifetime such as CO.

AS09-A053
Trends of Ozone and Particulate Matter in a Metropolitan City of Northeastern Pennsylvania.

Brandon WHITMAN1+, Saritha KARNAE2#
1Wilkes University, United States, 2Department of Environmental Engineering and Earth Sciences, United States


Epidemiological studies conducted over past decade have provided evidence on adverse health effects in sensitive groups (young children and elderly) and adults upon exposure to elevated fine (PM2.5) and coarse particulate matter (PM10) concentrations. In addition to particulate matter studies have also concluded effects of ozone, which is precursor of particulate matter on human health and welfare. In northeastern Pennsylvania Lackawanna county is the second-largest county within the metropolitan area while Luzerne County is the second-largest by total area. These counties are home to Scranton-Wilkes-Barre-Hazleton Metropolitan Statistical Area.  Continuous ambient monitoring stations (CAMS) have been set up by the Pennsylvania Department of Environmental Protection agency in Scranton and Wilkes-Barre to measure particulate matter and ozone. A decrease in the design values of ozone was noted in the metropolitan area however design values measured in Wilkes-Barre were noted to be higher than Scranton. Ozone is precursor of secondary particulate matter thus the current study focus on evaluation of statistical trends both long-term and seasonal. Trajectory and cluster analysis will also be used to identify the potential local vs long-range transport from surrounding states.

AS12-A002
3D Precipitation Nowcasting: Resnet Applied to Highly Dense Pawr Data

Maha MDINI1#+, Takemasa MIYOSHI1,2, Shigenori OTSUKA1
1RIKEN Center for Computational Science, Japan, 2University of Maryland, United States


Sudden heavy rain may lead to disasters like flooding and loss of life and property. To reduce the risk, predicting sudden downpours is of key importance. However, predictability of such events is limited to only for a very short range within an hour or shorter because of their abruptness. In this case nowcasting is an effective approach. Detecting sudden heavy rain even 10 minutes before it occurs can reduce the damage drastically. Precipitation nowcasting is the process of short-range prediction based on observation data. In the case of sudden rainfalls, this process is difficult due to the fast evolution of the rain and its chaotic nature. Therefore, we need innovative techniques. The novel Phased-Array Weather Radar (PAWR) offers dense 3D images of reflectivity every 30 seconds. We took advantage of this big data to perform nowcasting using neural networks. We use Residual Neural Networks (RESNet) to compress the images and extract information relevant for the prediction. Next, we use a Convolutional Neural Network (CNN) to make the prediction. Afterwards, we use the same RESNet to map the forecast to the original domain. The RESNet and the CNN are trained jointly for thecompression to maximize the prediction accuracy. Our first results show that in most cases we can predict precipitations up to 30 minutes, with an error rate (false positives + false negatives) of 8%. The use of the RESNet allowed to alleviate the memory load and the computational complexity of the prediction. Moreover, training the RESNet and the CNN jointly reduced immensely the prediction noise in non-precipitation regions and improved the accuracy in precipitation regions. 

AS12-A007
Cold Surge Effect on Atmospheric Sounding Parameter Over Natuna Island During East Asia Winter Monsoon

Wignyo PRASETYO1#, Yosafat Donni HARYANTO1+, Nelly Florida RIAMA2, Feriomex HUTAGALUNG2
1State College of Meteorology Climatology and Geophysics, Indonesia, 2Indonesian Agency for Meteorology, Climatology and Geophysics, Indonesia


During the East Asian Winter Monsoon from November to February, cold surge event often occur which result in increased convective activity in the western Indonesian Maritime Continent. Natuna Island, which is located in the southern part of the South China Sea, is important to study because it is directly affected by cold surge propagation. In this study, we analyzed the thermodynamic structure of the atmosphere based on radio sonde data at 00 and 12 UTC. The results obtained that air temperature and potential air temperature tend to be higher in the lower troposphere, but lower in the upper troposphere during a cold surge. Relative Humidity is higher during cold surges, a significant increase occurs in the upper troposphere. The wind direction of the lower troposphere changes from easterly to north-easterly with a higher speed than on days without cold surge. Lifting Condensation Level and Equillibrium Level were elevated, while Level of Free Convection were decreased although the difference is small. Precipitable water were increased in all observed troposphere layers. This is followed heavy rain intensity, especially if cold surges occur intensely even though the Convective Available Potential Energy were decreased and the Convective Inhibition were increased. In general, the occurrence of cold surges resulted a significant increase in atmospheric instability at 00 UTC compared to 12 UTC. Keywords: cold surge, LCL,CAPE 

AS03-A003 | Invited
Gravity Wave-induced Variations in Sodium Nightglow: Model Development and Initial Results

Tai-Yin HUANG1#+, Yanlin LI2, Yolian AMARO-RIVERA3, Fabio VARGAS4, Julio URBINA2
1Penn State Lehigh Valley, United States, 2The Pennsylvania State University, United States, 3Johns Hopkins University, United States, 4University of Illinois at Urbana-Champaign, United States


A two-dimensional nonlinear Na Chemistry Dynamics (NaCD) model has been developed to study Sodium nightglow variations induced by gravity waves.  The current chemistry used in the NaCD model has primarily taken from the Na chemistry from Plane (2014) but excluding the photochemical reactions.  We also included three reactions that were not considered in Plane (2014).  These three reactions involved O, H, and O3, which might play a significant role given that atomic oxygen is an important player in the MLT region.  The model simulations will output Na nightglow intensity and Na number densities which will be compared to the Na observations from the Na Lidar in the Andes Lidar Observatory.  We will also present a comparison of simulated wave-induced Na nightglow intensity variations to other nightglow variations in the airglow layers like OH, O2 atmospheric band, and O(1S) greenline.                

AS03-A002
Southern Hemisphere Response to Northern Hemisphere Stratospheric Warming Events

Patrick ESPY1,2#+, Kristina SHIZUKA YAMASE SKARVANG1, Willem VAN CASPEL1, Robert HIBBINS1,2, Andrew KAVANAGH3
1Norwegian University of Science and Technology, Norway, 2University of Bergen, Norway, 3British Antarctic Survey, United Kingdom


The British Antarctic Survey meteor radar located at Rothera station on the Antarctic Peninsula (68S, 68W), has been used to measure hourly winds from 82 km to 98 km with 2 km vertical resolution from February 2005 to August 2019. The mean winds, as well as the 24- 12- and 8-hour tidal amplitudes were fitted to these winds using a 4-day running window. To remove seasonal effects, an individual, smoothed climatology for each of the tidal components and the mean wind was constructed and used to calculate wind anomalies. A superposed-epoch analysis quantified the behaviour of the wind and tidal amplitude anomalies during the 5, northern hemisphere stratospheric warming events that were accompanied by an elevated stratopause occurring between 2005-2013. The results of that analysis showed that the northern hemisphere events did not cause any significant change in the mean winds or the 24- and 8-hour tidal amplitudes. However, the 12-hour tidal amplitude showed a decrease of between 1 m/s at 84km, and 11 m/s at 92 km, significant between 2 and 8 sigma, 12 to 15 days after the event. Preliminary results using the WACCM-X-SD model show similar southern hemisphere behaviour, and the cause of these anomalies is investigated using a primitive-equation tidal model incorporating mean wind fields derived from the NAVGEM-HA reanalysis. We will present the data, analysis, and model results showing the response of southern hemisphere winds to northern hemisphere stratospheric warming events.   

AS03-A012
Impact of the Major Ssws of February 2018 and January 2019 on the Middle Atmospheric Nitric Oxide Abundance

Yvan ORSOLINI1#+, Kristell PÉROT2
1Norwegian Institute for Air Research, Norway, 2Chalmers University of Technology, Sweden


The Arctic middle atmosphere was affected by major sudden stratospheric warmings (SSW) in February 2018 and January 2019, respectively. In this article, we report for the first time the impact of these two events on the middle atmospheric nitric oxide (NO) abundance. The study is based on measurements obtained during two dedicated observation campaigns, using the Sub-Millimetre Radiometer (SMR) aboard the Odin satellite, measuring NO globally since 2003.The SSW of February 2018 was similar to other, more dynamically quiet, Arctic winters in term of NO downward transport from the upper mesosphere-lower thermosphere to lower altitudes (referred to as energetic particle precipitation indirect effect EPP-IE). On the contrary, the event of January 2019 led to one of the strongest EPP-IE cases observed within the Odin operational period. Important positive NO anomalies were indeed observed in the lower mesosphere-upper stratosphere during the three months following the SSW onset, corresponding to NO volume mixing ratios more than 50 times higher than the climatological values. These different consequences on the middle atmospheric composition are explained by very different behavior of the polar stratopause during these two SSW events.

AS03-A009
Validation of Ssusi Derived Auroral Ionization Rates and Electron Densities

Stefan BENDER1#+, Patrick ESPY2,3, Larry PAXTON4
1Instituto de Astrofísica de Andalucía (CSIC), Spain, 2Norwegian University of Science and Technology, Norway, 3University of Bergen, Norway, 4Johns Hopkins University Applied Physics Laboratory, United States


Solar, auroral, and radiation belt electrons enter the atmosphere at polar regions leading to ionization and affecting its chemistry. Climate models usually parametrize this ionization and the related changes in chemistry based on satellite particle measurements. Precise measurements of the particle and energy influx into the upper atmosphere are difficult because they vary substantially in location and time. Widely used particle data are derived from the POES and GOES satellite measurements which provide electron and proton spectra. We validate the electron energy and flux data products from the Special Sensor Ultraviolet Spectrographic Imager (SSUSI) instruments on board the Defense Meteorological Satellite Program (DMSP) satellites. This formation of currently three operating satellites observes the auroral zone in the UV from which electron energies and fluxes are inferred in the range from 2 keV to 20 keV. We use these observed electron energies and fluxes to calculate ionization rates and electron densities in the lower thermosphere (≈ 90–150 km), and validate them against EISCAT ground-based measurements by comparing the calculated to observed electron density profiles. We find that with the current standard parametrizations, the SSUSI-derived auroral electron densities (90–150 km) agree well with the ground-based measured ones. Because of the large orbit-to-orbit variation, the differences are not significant.

AS03-A011
Unusual Quasi-10-day Waves and Ionospheric Responses During the 2019 Southern Hemisphere Sudden Stratospheric Warming

Jack WANG1#+, Scott PALO2, Jeffrey FORBES2, John MARINO2, Tracy MOFFAT-GRIFFIN3
1NASA Goddard Space Flight Center, United States, 2University of Colorado Boulder, United States, 3British Antarctic Survey, United Kingdom


An unusual sudden stratospheric warming (SSW) occurred in the Southern hemisphere in September 2019. Ground-based and satellite observations show the presence of transient eastward- and westward-propagating quasi-10 day planetary waves (Q10DWs) during the SSW. The planetary wave activity maximizes in the MLT region approximately 10 days after the SSW onset. Analysis indicates that the westward-propagating Q10DW with zonal wave number s = 1 is mainly symmetric about the equator, which is contrary to theory which predicts the presence of an antisymmetric normal mode for such planetary wave. Observations from MLS and SABER are combined with meteor radar wind measurements from Antarctica, providing a comprehensive view of Q10DW wave activity in the southern hemisphere during this SSW. Analysis suggests that the Q10DWs of various wavenumbers are potentially excited from nonlinear wave-wave interactions that also involve stationary planetary waves with s = 1 and s = 2. The Q10DWs are also found to couple the ionosphere with the neutral atmosphere. The timing of the quasi-10-day oscillations (Q10DOs) in the ionosphere are contemporaneous with the Q10DWs in the neutral atmosphere during a period of relatively low solar and geomagnetic activity, suggesting that the Q10DWs play a key role in driving the ionospheric Q10DOs during the Southern SSW event. This study provides observational evidence for coupling between the neutral atmosphere and ionosphere through the upward propagation of global scale planetary waves.

AS41-A020
Sensitivity Experiments of the Dry Dynamical Core Using Cwbgfs

Nai-Wen CHANG+, Chien-Ming WU#
National Taiwan University, Taiwan


The dynamical core plays an important role in the performance of an atmospheric numerical model, consisting of the equations that control the movement and state of the air. Since 2008, The Dynamical Core Model Intercomparison Project (DCMIP) has been held to test the dynamical cores among different numerical models with topics such as barely dry dynamic and non-hydrostatic dynamic. This study intends to conduct dry dynamical core tests on the Central Weather Bureau Global Forecast System (CWBGFS), following idealized cases including steady state and baroclinic wave in the DCMIP-2008 document (Jablonowski et al. 2008; Jablonowski and Williamson, 2006). The CWBGFS is a hydrostatic spectral model under a semi-implicit scheme, which has been developed by the Central Weather Bureau (CWB) in Taiwan since 1980. Recently, the CWBGFS has implemented a semi-Lagrangian scheme on octahedral reduced Gaussian grid (Malardel et al. 2016) with resolutions of 640 latitude circles from pole to equator and 72 vertical levels (TCo639L72, ~15km for horizontal resolution). We will perform sensitivity experiments using CWBGFS and be compare the results with the reference states provided by Jablonowski and Williamson (2006) to evaluate the performance of the CWBGFS.



AS20-A020 | Invited
Mesoscale Processes Associated with the Formation Stages of Explosive Extratropical Cyclogenesis in the North Pacific: 1979-2021

John GYAKUM#+
McGill University, Canada


Explosive extratropical cyclogenesis is generally accepted as primarily a cold-season, maritime process that has been the subject of extensive research during the past 40 years. Our objective is to document the predominantly mesoscale processes that facilitate the incipient stages of cyclones that ultimately deepen explosively during the North Pacific winter. In particular, we focus on the mesoscale environment that facilitates the formation of these cyclones, often as a relatively small-scale secondary development associated with cold frontogenesis from a much larger, and deeper cyclone. The mesoscale environment is characterized by poleward intrusions of tropical air masses. The consequences of such intrusions include a reduction in the effective static stability that increases the strength of the secondary circulation response to a given strength of forcing (such as mid-tropospheric cyclonic vorticity advection), in addition to the nonlinear feedbacks on precipitation rate for air masses characterized by higher equivalent potential temperatures at approximately 850 hPa. The temperature stratification above these levels of maximum equivalent-potential temperature is typically moist neutral, suggesting the crucial importance of elevated moist convective processes in cold-season extratropical cyclones. Documentation of the cyclogenesis processes is obtained with the 6-hourly manually analyzed surface maps of the Northern Hemisphere, along with the European Centre for Medium-Range Forecasting reanalysis (ERA5).By using the unique combination of this surface cyclone dataset and the high-resolution global reanalysis, we examine the modulation of explosive cyclogenetic processes by the weak stratification in the presence of surface frontogenetic forcing. In particular, we assess the role that warming air masses have in modulating the mesoscale processes associated with incipient phases of explosive cyclogenesis in the North Pacific basin. 

AS20-A005
Near-surface Strong Winds in Typhoon Faxai Observed by Saitama MP-PAWR in 2019

Shinsuke SATOH1#+, Fusako ISODA1, Hironori IWAI1, Hanado HIROSHI1, Katsuhiro NAKAGAWA1, Shigenori OTSUKA2, Takemasa MIYOSHI2,3, Takeshi MAESAKA4, Shingo SHIMIZU4
1National Institute of Information and Communications Technology, Japan, 2RIKEN Center for Computational Science, Japan, 3University of Maryland, United States, 4National Research Institute for Earth Science and Disaster Resilience, Japan


Typhoon Faxai, which crossed the Kanto region on September 9, 2019, caused severe damage in various places. In Chiba City, a maximum instantaneous wind speed of 57.5 ms-1 was recorded. In Chiba Prefecture, a steel tower collapsed and caused a large-scale power outage. The multi-parameter phased array weather radar (MP-PAWR) installed at Saitama University can observe only a part of the northwestern part of Chiba prefecture where the damage was large, but the view in that direction was well opened, so the Doppler velocities near the ground surface were observed. To investigate typhoon storm disasters, it is important to observe accumulated rainfall and strongest winds near the surface. However, it is often difficult to measure the wind speed near the surface using Doppler radar due to obstacles and the curvature of the earth. Also, one radar can only a wind component in the radial direction. The wind speed measured by Doppler radar is the spatial average within the observation volume, even if the observation time is less than 0.1 seconds. However, observations of the Doppler velocity width, which indicates velocity variations in the observation volume, may be useful for estimating gust rates. In this study, we aim to estimate the near-surface winds using the MP-PAWR fine 3D data in 100 elevation angles observed every 30 seconds. According to the observation data, the strongest wind reached near the surface near the typhoon center. In the strongest wind region, the vertical shear and the region where the Doppler velocity width exceeded 3 ms-1 extended to a height of 2 km or less.

AS20-A016
A Modelling Study of Possible Impacts of Future Climate Change on Strong Typhoons in the Western North Pacific

Yi Ting THEAN#+, Chung-Chieh WANG, Zhe-Wen ZHENG
National Taiwan Normal University, Taiwan


A demand of future assessment to an environment influenced by greenhouse gases (GHG) effect is an appetence for scientists. This spatial and temporal traversing experiment was designed under actual scenario from six recent tropical cyclone (TC) events with two Representative Concentration Pathway (RCP) future scenarios by 38 Coupled Model Intercomparison Project (CMIP) phase five models. The most intense Western North Pacific Ocean (WNPO) TCs from 21st century onwards were selected for this study. Fifty Cloud Resolving Storm Simulator (CReSS) simulations were simulated under 2.5 km horizontal resolution. Most TCs resulted weaker intensity in RCP experiments, while sensitivity experiments in three most significant cases has demonstrated impacts by the following possible reasons: differences in relative humidity, initial vortex strength, and environmental vertical wind shear.

AS20-A009
Aircraft Observations of Intense Typhoons in 2017 and 2018 in the T-PARCII Project

Kazuhisa TSUBOKI1#+, Hiroyuki YAMADA2, Tadayasu OHIGASHI3, Taro SHINODA1, Kosuke ITO4, Munehiko YAMAGUCHI5, Tetsuo NAKAZAWA6, Hisayuki KUBOTA7, Yukihiro TAKAHASHI7, Nobuhiro TAKAHASHI1, Norio NAGAHAMA8, Kensaku SHIMIZU8
1Nagoya University, Japan, 2University of the Ryukyus, Japan, 3National Research Institute for Earth Science and Disaster Resilience, Japan, 4Kyoto University, Japan, 5Japan Meteorological Agency, Japan, 6The University of Tokyo, Japan, 7Hokkaido University, Japan, 8Meisei Electric, Japan


Historical data of typhoon (best track data) include large uncertainty after the US aircraft reconnaissance of typhoon was terminated in 1987. To improve this problem, in situ observations of typhoon using an aircraft are indispensable. The T-PARCII (Tropical cyclone-Pacific Asian Research Campaign for Improvement of Intensity estimations/forecasts) project is aiming to improve estimations and forecasts of typhoon intensity as well as storm track forecasts. In 2017, the T-PARCII team performed dropsonde observations of Supertyphoon Lan in collaboration with Taiwan DOTSTAR, which was the most intense typhoon in 2017 and caused huge disaster over the central Japan. Lan moved northeastward to the east of the Okinawa main island and it was located around 23 N on 21 and 28 N on 22 October. In these two days, we made dropsonde observations at the center of the eye and in the surrounding area of the eyewall. The observations showed that the central pressure of Lan slightly increases from 926 hPa on 21 to 928 hPa on 22 October with the northward movement. The observations showed a significant double warm core structure in the eye and the maximum wind speed along the eyewall. The T-PARCII team also made aircraft observations of Typhoon Trami during the period from 25 to 28 September 2018 in collaboration with the SATREPS ULAT group and DOTSTAR. Trami was almost stationary during the period to the south of the Okinawa main island. Then, it moved northward and finally made a landfall over the central part of Japan. Typhoon Trami showed a drastic change of intensity from 25 to 26 September with a large change of eye size from about a diameter of 60 km to 200 km. Dropsonde observations showed the change of central pressure and maximum wind speed as well as the thermodynamic structure of the eye.

AS20-A014
Origin of Water Vapor and Structure of an Atmospheric River in the Heavy Rain Event of July 2018

Yunhee KANG1+, Kazuhisa TSUBOKI1#, Masaya KATO1, Jonghoon JEONG2
1Nagoya University, Japan, 2Joint Institute for Regional Earth System Science and Engineering (JIFRESSE), University of California Los Angeles, United States


The heavy rain event in July 2018 caused an enormous socioeconomic impact in western Japan, officially named as the “Heavy Rain Event of July 2018” by the Japan Meteorological Agency (JMA). During this event, a persistent corridor of strong water vapor formed from the East China Sea (ECS) to the Japanese Islands that is the so-called atmospheric river (AR). Under this environment, linear mesoscale convective systems (MCSs), interacted with terrain, caused the heavy rainfalls and flooding events in western Japan. The present study investigated the origin of water vapor and the structure of the AR that cause the Heavy Rainfall Event of July 2018. The anomalous northeast retreat and weakened intensity of the North Pacific Subtropical High (NPSH), and the passage of Typhoon Prapiroon (2018) influenced the formation of the strong water vapor corridor. The AR had a length of approximately 3000 km from the ECS to Japan and persisted for about 24 hours. Backward trajectories showed that the primary source region of the warm and moist air entering the AR was from the South China Sea. The feeder-stream around Taiwan also contributed to the maintenance of the AR by transporting moisture in the low-level.  

AS20-A012
Impacts of Vertical Profiles of Water Vapor in the Upstream Region on the Heavy Rainfall Event of July 2018 in Japan

Taro SHINODA#+, Kamiya ASUKA, Masaya KATO, Sachie KANADA, Kazuhisa TSUBOKI
Nagoya University, Japan


Vertical profiles of water vapor have a crucial role to the rainfall amount. A heavy rainfall event is brought about in the western Japan in the beginning of July 2018. A previous study pointed out that the convective activity in the upstream region over the East China Sea where approximately 1000 km far from the heavy rainfall region should be moistening the middle troposphere and a deep moist air-mass intrusion should have an important role in the event. This study explores impacts of vertical profiles of water vapor in the upstream region on the rainfall amount in the western Japan using a Cloud Resolving Storm Simulator (CReSS). The simulation captures well the rainfall amount and its horizontal distribution in the western Japan. Trajectory analyses shows that the air-mass intruding into the western Japan has its origin in the upstream region. The sensitivity experiments which increase (decrease) water vapor amount in the lower troposphere in the upstream region show clearly the increasing (decreasing) rainfall amount in the western Japan. On the other hand, no significant change in the rainfall amount in the western Japan appears in sensitivity experiments which changing water vapor amount in the middle troposphere. In the drier experiments, convective activities along the trajectory are moistening the middle troposphere. In the moister experiments, convective activities along the trajectory remove water vapor in the middle troposphere by convective precipitation, and rainfall regions shift into the upstream side. As a result, precipitable water amounts both in drier and moister experiments are approximately equal in the western Japan. The results show that the water vapor observation, especially in the lower troposphere, in the upstream region should be quite important to the forecasting of the rainfall distribution and amount.

AS20-A010
Evolution of a Convective Storm Revealed by High Temporal-spatial Resolution Observation of Precipitation Cores

Yukie MORODA1#+, Kazuhisa TSUBOKI1, Shinsuke SATOH2, Katsuhiro NAKAGAWA2, Tomoo USHIO3, Shingo SHIMIZU4
1Nagoya University, Japan, 2National Institute of Information and Communications Technology, Japan, 3Osaka University, Japan, 4National Research Institute for Earth Science and Disaster Resilience, Japan


An isolated convective storm was observed by two phased array weather radars in the Kinki District, western Japan on 7 August 2015. Using their rapid scanning capability which can complete one volume scan in 30 seconds, we investigated the evolution of the convective storm in detail. To describe the convective storm’s evolution, we used the following definitions. The precipitation cell is defined as a three-dimensionally contiguous region of 40 dBZ or greater, and the precipitation core is defined by a threshold of positive deviation greater than 7 dBZ from the average reflectivity during the mature stage of the cell. An updraft core is defined as an updraft region of 1 m/s or stronger at a height of 2 km.The storm was judged as a single cell according to the above definition. During the mature stage of the cell, we identified nine precipitation cores and five updraft cores. A long-lasting updraft core and its branches moved southwestward or southeastward. Around these updraft cores, the precipitation cores were generated successively and three precipitation cores were generated simultaneously within a few minutes. The updraft core with the longest duration lasted 73.5 minutes. In contrast, the lifetimes of precipitation cores were from 4.5 to 14.5 minutes. The precipitation cell was maintained by the successive generations of updraft cores lifting low-level humid air. The total amounts of water vapor inflow supplied by all the identified updraft cores were proportional to the volumes of the precipitation cell, with a correlation coefficient of 0.75. The structure and evolution of the storm were described using precipitation cores and updraft cores that were clearly defined with threshold values of radar reflectivity and vertical velocity. We found that precipitation cores and updraft cores had different evolution characteristics in the storm.



AS43-A015 | Invited
Aviation Turbulence Forecasting: Review and Suggestions for Future Research Needs

Robert SHARMAN#+
National Center for Atmospheric Research, United States


The prediction of turbulence for aviation applications has a relatively short history, beginning in the 1950s with clear-air turbulence (CAT) encounters of long haul commercial airline flights at upper levels.  Although pilots were familiar with turbulence at lower levels from the dawn of aviation, CAT encounters, especially were a new phenomenon, and rules of thumb provided the only guidance for aviation turbulence forecasters, and these were often difficult to invoke and were based on incomplete understandings of the genesis and life cycles of turbulence. This lead to the establishment of a pilot reporting system (PIREPs) to identify encountered turbulence areas, a system which is still in place today. Over the years, especially the last 20 or so, our understanding of aviation turbulence has increased, in part because of the enhanced network of observations available, both ground-based and airborne. Given that aircraft respond to disturbances most favorably at scales about the size of the aircraft, direct prediction is not possible today, or even in the foreseeable future. However, progress has been made in turbulence forecasting based on ensemble approaches.  Further, the implementation of finer and finer routine operational numerical weather prediction (NWP) models has allowed better inferences of aviation-scale turbulence likelihood, and this will continue to proceed rapidly as NWP models continue to better predict small-scale transient phenomena such as thunderstorms. The use of higher-resolution simulations to reproduce observed cases of turbulence have also contributed to our understanding of turbulence over the last few years, and this enhanced understanding should help in suggesting better diagnostics of turbulence. This talk will briefly review the unique aspects of the aviation turbulence forecasting problem including observation requirements, summarize accomplishments to date, and suggest future research and operational and dissemination directions. 

AS43-A017
Probabilistic Forecasting of Upper Level Aviation Turbulence Using Multi-parameter and Time-lagged Ensembles

Hyeyum (Hailey) SHIN#+, Wiebke DEIERLING, Domingo MUNOZ-ESPARZA, Robert SHARMAN
National Center for Atmospheric Research, United States


Today’s operational aviation turbulence forecasts are based on operational numerical weather prediction (NWP) models, which cannot resolve aircraft-scale turbulence [O(100 m) or smaller]. Operational turbulence forecasts rely on turbulence diagnostics, which link resolved large-scale features responsible for generation of aircraft-scale turbulence [O(1-10 km) or larger] to unresolved small-scale turbulence (Sharman and Pearson 2017). Therefore, operational turbulence forecasts are impacted by the uncertainty in turbulence diagnostics, in addition to the uncertainty in large-scale weather forecasts from NWP models. In efforts to represent these uncertainties recent studies have applied probabilistic forecasting to turbulence forecasts, by generating either ensembles of possible weather states or ensembles of turbulence diagnostics for a deterministic weather state. In this study, we provide a comprehensive comparison of various methods of producing probabilistic forecasts based on a number of operational global NWP models (e.g., NCEP GFS) and a number of turbulence diagnostics. Our initial comparison focused on single-model based methods: time-lagged ensemble (Xu et al. 2019) tied to the uncertainty in weather forecasts, multi-parameter ensemble (Kim et al. 2018) associated with the uncertainty in turbulence diagnostics, and a combination of them. From case studies, it was found that time-lagged ensembles cannot produce a sufficiently large spread, due to the small number of ensemble members available within a forecast length constraint of 36 hours, while the multi-parameter ensembles produce a relatively large spread. Objective evaluations using in-situ eddy dissipation rate observations for the 1-month period of December 2019 showed that the multi-parameter method outperforms the time-lagged method. The blended method showed diverse results, depending on the choice of time-lagged ensemble members. This research is in response to requirements and funding by the Federal Aviation Administration (FAA). The views expressed are those of the authors and do not necessarily represent the official policy or position of the FAA. 

AS43-A018
Development of the Global Korean Aviation Turbulence Guidance (G-KTG) System Based on the KMA’s Operational Unified Model and Its Evaluation Using the Global Turbulence Observation

Dan-Bi LEE1+, Hye-Yeong CHUN1#, Soo-Hyun KIM2, Robert SHARMAN3, Jung-Hoon KIM2
1Yonsei University, Korea, South, 2Seoul National University, Korea, South, 3National Center for Atmospheric Research, United States


The Global-Korean aviation Turbulence Guidance (G-KTG) system is developed using the Unified Model (UM)-based Global Data Assimilation and Prediction System (GDAPS) of Korea Meteorological Administration (KMA) with a resolution of 10 km. The final G-KTG product is calculated by ensemble means of the Clear-Air Turbulence (CAT), Near-Cloud Turbulence (NCT), and Mountain-Wave Turbulence (MWT) diagnostics. For the CAT forecast, the diagnostics in the Graphical Turbulence Guidance version 3 (GTG3) and KTG systems are used separately. The G-KTG 12-h forecasts for three altitude regions are evaluated against the eddy dissipation rate observation for a 1-yr period recorded from the commercial flight, using the statistical skill scores based on the Probability Of Detection (POD) method. Among various combinations between the CAT, NCT, and MWT diagnostics, the combination of the CAT and NCT diagnostics at upper levels and that of the CAT and MWT diagnostics at mid- and low-levels have the highest Area Under Curve (AUC) value. In terms of the hit rate and true skill statistic, the combination of the CAT, MWT, and NCT diagnostics provides the best performance at upper-levels. According to this, we suggest that the G-KTG combining the CAT, MWT, and NCT diagnostics will be helpful for averting possible turbulence encounters. In the seasonal performance tests, all combinations of the G-KTG show the best performance in either fall or winter, while the lowest performance in summer. It is also shown that the G-KTG including the NCT diagnostics significantly improves the performance in the summer. In the comparison of performance between two G-KTGs based on the GTG3 and KTG, the GTG3-based G-KTG shows higher AUC in both US (0.83) and East Asia (0.81) regions than the KTG-based G-KTG (US: 0.81 and EA: 0.78).

AS43-A021
A Case Study for Convectively Induced Turbulence Using In Situ Aircraft, Radar Spectral Width Data, and WRF Simulation

Jung-Hoon KIM#+
Seoul National University, Korea, South


A commercial aircraft, departing from Seoul to Jeju Island in South Korea, encountered a convectively induced turbulence (CIT) at about z = 2.2 km near Seoul on 28 October 2018. At this time, the observed radar reflectivity showed that the convective band with cloud tops of z = 6–7 km passed the CIT region with high values of spectral width (SW; larger than 4 m s–1). Using the 1 Hz wind data recorded by the aircraft, we estimated an objective intensity of the CIT as a cube root of eddy dissipation rate (EDR) based on the inertial range technique, which was about 0.33–0.37 m2/3 s-1. Adar-based EDR was also derived by the EDR was about 0.3–0.35 m2/3 s-1 near the CIT location, which is consistent with in situ EDR. In addition, a feasibility of the CIT forecast was tested using the weather and research forecast (WRF) model with a 3 km horizontal grid spacing. The model accurately reproduced the convective band passing the CIT event with an hour delay, which allows the use of two methods to calculate EDR: The first is using both the sub-grid and resolved turbulent kinetic energy to infer the EDR; the second is using the LMT for converting absolute vertical velocity (and its combination with the Richardson number) to EDR-scale. As a result, we found that the model-based EDRs were about 0.3–0.4 m2/3 s-1 near the CIT event, which is consistent with the estimated EDRs from both aircraft and radar observations. This work was funded by the Korea Meteorological Administration Research and Development Program under grant KMI2020-01910, and was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2019R1I1A2A01060035). 

AS43-A003 | Invited
Aviation Meteorology in a Changing Climate (Invited)

Paul WILLIAMS#+
University of Reading, United Kingdom


Climate change has important consequences for aviation, because the atmosphere’s meteorological characteristics strongly influence flight routes, journey times, and turbulence. For example, a significant amount of aircraft turbulence is generated by vertical wind shear. New evidence shows that the jet stream has become 15% more vertically sheared at aircraft cruising altitudes since satellites began observing it in 1979. To investigate the consequences for aviation, we diagnose a basket of clear-air turbulence measures from climate model simulations. We find that turbulence strengthens significantly under climate change, all around the world, in all seasons, and at a wide range of aircraft cruising altitudes. For example, within the transatlantic flight corridor in winter at around 39,000 feet, the occurrence of light turbulence increases by 43–68%, moderate by 37–118%, and severe by 36–188%. These findings underline the urgent need to improve the skill of operational clear-air turbulence forecasts, to avoid increases in on-board discomfort and injuries in the coming decades. To investigate the influence of climate change on flight routes and journey times, we feed atmospheric wind fields generated from climate model simulations into a flight routing algorithm. We focus on transatlantic flights between London and New York, and how they change when the CO2 concentration is doubled. We find that a strengthening of the prevailing jet-stream winds causes round-trip journeys to significantly lengthen in all seasons. Extrapolation of our results to all transatlantic traffic suggests that aircraft will collectively be airborne for an extra 2,000 hours each year, burning an extra 7.2 million gallons of jet fuel at a cost of US$ 22 million, and emitting an extra 70 million kg CO2. The above findings provide further evidence of the two-way interaction between aviation and climate change, which is an emerging research area that deserves further study.

AS43-A007
Global Response of Near Cloud Aviation Turbulence to Climate Change

Soo-Hyun KIM1#+, Jung-Hoon KIM1, Hye-Yeong CHUN2
1Seoul National University, Korea, South, 2Yonsei University, Korea, South


Encounters with turbulence in flight are a major hazard to the aircraft operation that is responsible for passenger and crew discomfort, structural damages, and extra fuel consumptions. Among various potential sources of turbulence, convection is one of well-known sources which causes in-cloud convectively induced turbulence (CIT) and out-of-cloud CIT [or near-cloud aviation turbulence (NCT)]. Although previous studies using climate models showed that climate change can bring turbulence intensification, which leads to bumpier flights, these turbulence diagnoses were limited to clear air turbulence generation. In this regard, we investigate effects of climate change to NCT generation using climate models. We use 45 years of data for future period (2056-2100) with the shared socio-economic pathway 5-8.5 emission scenario and 45 years (1970-2014) of historical data from historical simulation. Here, the NCT is diagnosed using turbulence diagnostics developed based on the convective gravity wave drag parameterization scheme of Chun and Baik. The convective heating rate, which is required for calculation of the NCT diagnostics but is not provided from climate models, is estimated using existing algorithm included in the Weather Research and Forecasting model’s Read/Interpolate/Plot package, which uses air temperature, water vapor mixing ratio, and vertical velocity. For both future and historical periods, the probability density functions of the NCT diagnostics and the 95th percentile as a threshold of moderate-or-greater (MOG) turbulence are calculated. Finally, we calculate the occurrence frequency of the MOG turbulence for both periods separately and analyze changes to NCT over the entire globe by comparing global occurrences between two periods. Sensitivity of turbulence occurrences on climate models is also investigated. The results will be presented in the conference. Acknowledgement: This research is supported by the Korean Meteorological Administration Research and Development Program (KMI2020-01910), and by the National Research Foundation of Korea Research and Development Program (NRF-2019R1I1A2A1060035).



AS24-A004 | Invited
Regulation of the Subseasonal Variability of Winter Rainfall in South China by the Diversity of El Niño Southern Oscillation

Congwen ZHU#+, Li GUO, Boqi LIU
Chinese Academy of Meteorological Sciences, China


Winter precipitation over South China tends to increase with enhancement of the 10–30-day intraseasonal oscillation (ISO) during El Niño Southern Oscillation (ENSO) events from 1981 to 2017. Our results show that, in contrast with central Pacific El Niño and La Niña events, the 10–30-day ISO of rainfall particularly intensify during eastern Pacific (EP) El Niño events. The seasonal evolution of the anomalous circulation, indicated as the annual cycle (AC), bridges the slow-varying ENSO and the transient 10–30-day ISO of winter rainfall over South China. As to the AC component, the EP El Niño events not only enhance the Philippine Sea anticyclone to provide a wetter low-level background over South China, but develop the mid-latitude cyclone and westerly winds in the upper-level over East Asia by changing tropical convection. The moisture over South China further increases due to the low-level wind convergence on a 10–30-day timescale. In the upper troposphere, the AC component of the anomalous westerly redistributes the 10–30-day relative vorticity by modulating the subseasonal mid-latitude wave train to strengthen the positive vorticity advection and ascending motion over South China. Their collaboration amplifies the subseasonal variance of winter precipitation in situ. By contrast, the AC component of the upper level circulation at mid-latitudes is not well organized in other ENSO subsets, corresponding to their lack of influence on the ISO variability of rainfall over South China.

AS24-A002
Effects of ENSO Diversity and Cold Tongue Bias on Seasonal Prediction of South China Late Spring Rainfall

Ronald Kwan Kit LI1#+, Francis Chi Yung TAM1,2, Gabriel LAU1
1The Chinese University of Hong Kong, Hong Kong SAR, 2Shenzhen Research Institute, China


Seasonal prediction of South China April to May rainfall is examined based on forecasts by the European Centre for Medium-Range Weather Forecasts (ECMWF) operational model. El-Nin ̃o Southern Oscillation (ENSO) is a major source of predictability, conveyed by the lower tropospheric anomalous western Pacific anticyclone and cyclone for El-Nin ̃o and La-Nin ̃a respectively. By separating ENSO into its diversity of eastern Pacific (EP) and central Pacific (CP) ENSO, different effects on South China rainfall are revealed. From observations, while rainfall is enhanced following EP El-Nin ̃o and reduced following EP La-Nin ̃a, rainfall remains close to climatology following CP ENSO. However, the seasonal forecast model predicts CP ENSO effect on South China rainfall to be similar to EP ENSO. The model develops excessive westward extension of the eastern Pacific cold tongue within seasonal timescale. While model predicts tropical central Pacific anomalous sea surface temperature (SST) following CP ENSO realistically, the colder mean state is proposed to weaken the anomalous convection, which subsequently induces bias in the western Pacific anomalous rotational flow and hinders South China rainfall prediction. Meanwhile following EP ENSO, the colder mean state is proposed to strengthen the thermocline feedback, inducing stronger tropical eastern Pacific anomalous SST. While bias in the western Pacific anomalous rotational flow is also induced, the bias is far away from South China so rainfall prediction is still realistic. This study highlights the importance of model mean state in the fidelity of model ENSO diversity teleconnections on seasonal timescale.

AS24-A005
An Evaluation of the Predictability of the Subseasonal Peak Rainfall Event Associated with the Northern Queensland Floods of February 2019

Wayne Yuan-huai TSAI+, Mong-Ming LU#, Chung-Hsiung SUI
National Taiwan University, Taiwan


The northern Queensland floods of February 2019 has been studied to evaluate the multiple large-scale atmospheric influence on the S2S prediction skill of a high-impact extreme event. A subseasonal peak event, which was defined as the period with maximum 15-day accumulated rainfall amount within the boreal winter half year from November to April (NDJFMA) based on the CMORPH dataset, is chosen as the target for evaluating S2S prediction. The peak event is the most important wet episode of time scale within the range of 10~60 days, during which the accumulated rainfall shows distinctly contribution to the seasonal totals. The rainfall episode associated with the 2019 Queensland flood coincides with the subseasonal peak event. The positive skew forecast of the Queensland rainfall during the 15 days of the peak event can be predicted 20 days before the occurrence of the event. The influence of the Australia monsoon evolution, MJO, equatorial Rossby wave and southern hemisphere extratropical Rossby wave activity flux on the Queensland floods will be discussed using both observational and S2S prediction database.

AS24-A006
Development of Multimodel Ensemble Seasonal Climate Forecasting System Customized at Regional Scale

Soo-Jin SOHN#+
APEC Climate Center, Korea, South


In order to produce, distribute and utilize regional-specific climate forecasting data for the widespread use of seasonal climate data in consideration of a public service perspective in the Asia-Pacific region, high-quality climate forecasting data should be utilized in regional-specific seasonal climate forecasting. In order to evaluate the predictive performance of each participant model and to construct a differentiated MME, the ENSO complexity and its teleconnection were used as a physical factor for diagnosing and improving the predictive performance, and the model was selected based on this. The difference between the method used in this study and the existing methods is that ENSO is considered to be a combination of two major modes of tropical Pacific sea surface temperature, and the model was selected by evaluating the predictability of these two modes and their teleconnection. The effect of model selection was dependent on season and region, with better predictability compared to the original MME in East Asia and Australia summer precipitation forecasts.

AS24-A007
Improving Subseasonal Dynamic Forecast Over Korean Peninsula Using Deep Learning

Uran CHUNG#+, Kyoungwon PARK, Soo-Jin SOHN
APEC Climate Center, Korea, South


Demands for seamless information for mid-term (3~6 weeks) forecasts increase recently and how to improve S2S forecasting performance, deep learning techniques, which is recently in the spotlight is being introduced as for big data and artificial intelligence analysis techniques, are being introduced to improve S2S prediction. Therefore, we conducted this study to enhance the S2S prediction performance on the Korean Peninsula and to develop S2S predictive-based technologies that can be operated and utilized for the field. First, MME (multimodel ensemble)-based seasonal predicted climate data were constructed from six individual climate models (ECMWF, MSC, NCEP, KMA, UKMO, CMA), and the predicted performance was evaluated by learning with Long-Term Short Memory (LSTM) and Convolutional LSTM. In the results, RMSE and TCC were significantly improved in all three predicted climate variables (daily maximum and minimum temperature, and daily total precipitation) of MME-based S2S. In particular, in 3-4 week of the target leadtime, the TCC improved to 39% and 54% for the daily maximum temperature, and increased to 29% and 51% for the daily minimum temperature, respectively. TCC also improved dramatically in three week, to 130% for precipitation. Therefore, the application of deep learning training to the constructed MME-based S2S predicted climate data is expected to improve the S2S forecast climate for 3-4 weeks.

AS24-A009
Summer Prediction of the Western North Pacific Subtropical High and Yangtze River Rainfall Without Strong Enso Forcing

Chaofan LI1#+, Riyu LU1, Nick DUNSTONE2, Adam SCAIFE3,4, Philip BETT3, Fei ZHENG1
1Chinese Academy of Sciences, China, 2Met Office Hadley Centre, United Kingdom, 3Met Office, United Kingdom, 4University of Exeter, United Kingdom


During summer 2020, the Yangtze River basin suffered from extreme Meiyu rainfall, which broke the record since 1954 and led to catastrophic flood. This exceptional summer associated with anomalous change of the western North Pacific subtropical high (WNPSH), but without a strong forcing from ENSO. This study focuses on the seasonal prediction of the WNPSH during neutral summers without strong ENSO forcing, explores the associated predictable sources, and further diagnoses the seasonal predictability and associated dynamical causes for the unexpected Yangtze River basin rainfall in summer 2020. The results indicate that the coupled models show considerable prediction skill for the WNPSH during neutral summers, with successful reproduction of the WNPSH in the majority of neutral summers. The anomalous local WNP SST has an active role in modulating the variation of the WNPSH during neutral summers, could act as the primary predictable sources. For the exceptional Yangtze River rainfall in summer 2020, a good forecast of the anomalous WNPSH is presented, which contributes to a successful forecast of the above-average rainfall over the Yangtze River basin. The local WNP SST anomalies, in association with the Indian Ocean warming SST, acting as the major predictable tropical signals, are responsible for the westward extension of the WNPSH. However, the anomalous WNPSH couldn't totally explain the exceptional 2020 rainfall, as the forecast Yangtze River rainfall was much weaker comparable to that of observations. Our further results suggest that the evident enhancement of the EAJ intensity in summer 2020, gave a large boost to the exceptional Yangtze River rainfall, which are couldn't well reproduced by the models, hindering a better forecast of the intensity of the event and disaster mitigation.  

AS24-A015
Impact of Equatorial Rossby Waves and BSISO on East Asian Precipitation During the Boreal Summer

Qitong HUANG1#+, Shuguang WANG2, Jianping TANG1
1School of Atmospheric Sciences, Nanjing University, China, 2Nanjing University, China


During boreal summer seasons, plenty of tropical waves propagate northward and significantly influence East Asian surface weather in the midliatutdes. Among them, the Boreal Summer Intraseasonal Oscillations (BSISO) and Equatorial Rossby (ER) wave are two prominent wave phenomena active in the Indo-Pacific regions. BSISO and ER share similar intraseasonal time scales but display distinct propagating directions: BSISO propagates northeastward while ER propagates northwestward. Both ER and BSISO can influence East Asia weather as they propagate northward during the boreal summer seasons. The present study examine their impact on precipitation based on observational datasets. A BSISO index and an ERW index are used to composite precipitation anomalies and extreme precipitation on various BSISO and ERW phases. As the ERW phase progresses from west Pacific Ocean to Southeast Asia, precipitation anomalies and extreme precipitation shift accordingly. The impact of ERW can be observed in the hinterland of the south of China and west Pacific Ocean. On the other hand, the impact of BSISO on precipitation is most significant on the ocean, and its impact region ranges from 15N to 25N over the East Asia. The results over land are broadly consistent using both long term surface weather station measurements and satellite precipitation datasets.