Trigger mechanism of a snow burst event in Northeast China

Baofeng Jiao ,Lingkun Ran ,Xinyong Shen ,Yanin Qi

a Key Laboratory of Meteorological Disaster, Ministry of Education/ Joint International Research Laboratory of Climate and Environment Change/ Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing, China

b Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China

c Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai, China

d Jilin Weather Modification Office, Changchun, China

Keywords:Snow burst Frontogenesis Q -vector

ABSTRACT A snow burst event characterized by brief heavy snowfall affected Northeast China and caused serious social impact on 26 January 2017, with the snowband generally aligned with a northeast–southwest-oriented cold front.ECMWF reanalysis data were used to diagnose the possible trigger mechanism.Results showed there were two stages: (a) an initial stage far away from the Changbai Mountains, and (b) an enhancement stage under the influence of high terrain.During the initial stage, the coupling of low-level frontogenesis and a favorable convergence pattern caused strong upward motion, contributing to the release of instability.When the snowband approached the high terrain during the enhancement stage, the various instabilities were triggered by the low-level frontogenesis, terrain circulation, and strong wind shear associated with the low-level jet.Further, a modified Q -vector divergence including generalized potential temperature was calculated to diagnose the vertical motion.It showed that the frontogenesis terms contributed greatly to the negative Q -vector divergence along the moist isentropes,while the pseudo-vorticity terms played a role in the regions with strong wind shear associated with the low-level jet in the warm section, suggesting both were important in stimulating the ascending motion.The regions with negative Q -vector divergence had a close relationship with the vertical structure of convection, indicating the potential to track the development of the snowband in the next few hours.

1.Introduction

Heavy snow is one of the major catastrophic weather types during cold season.Snow events characterized by wide ranges, long durations, and high precipitation have been extensively examined( Bennetts and Hoskins, 1979 ; Sanders and Bosart, 1985 ; Tao and Wei,2008 ; Wang et al., 2009 ), while those of short duration and low precipitation, which may sometimes cause serious social impacts, have been less well studied ( Homan and Uccellini, 1987 ).Mesoscale convective systems related to the latter are referred to as snow squalls or snow bursts ( Milrad et al., 2011 ; Banacos et al., 2014 ).

In the past few decades, more attention has been payed to snow bursts.DeVoir (2004) noted this kind of high impact event accompanied by a sudden drop in temperature and increase in surface wind.The potential impacts and problems in terms of operational forecasting have been particular focused on.Subsequently, Pettegrew et al.(2009) performed a synoptic and mesoscale analysis on such a severe winter convection line.Detailed inspection revealed the line was strongly forced,triggered via persistent frontogenesis and various instabilities associated with the cold front.Several other case studies have been carried out based on analyses of mesoscale snowbands within synoptic cyclones,which have mainly served to highlight the influence of orographic forcing and passage of the cold front ( Schumacher et al., 2010,2015 ;Milrad et al., 2011 ).In addition to case studies, Lundstedt (1993) proposed a wintertime instability index based on moisture, instability, and lift.The index performed well in forecasting the potential for snow bursts in some cases.Banacos et al.(2014) proposed another tool that focused on near-surface moisture, instability, winds, and suitable temperatures to assess the occurrence of such events.With ascending motion assessed independently by surface isallobaric couplets and low-level frontogenesis, the tool was proved to have potential in highlighting favorable areas for snow bursts before in-depth assessment.

Previous research has shown the cold front and terrain to be the main factors responsible for snow initiation and enhancement.Frontal snow burst events in China take place frequently and can cause serous impacts.During the daytime of 26 January 2017 (UTC; LST = UTC + 8 h),a narrow linear convection system associated with a cold front moved swiftly through Northeast China, producing brief but heavy snow.Seven fatalities and two injuries were reported, along with the complete closure of some of the busiest expressways during the Spring Festival travel rush.Jiao et al.(2020) studied this event and focused on the instabilities responsible for snowband initiation and maintenance.However,it is unclear how the lifting was generated and what factors affected it.This paper focuses on the trigger mechanism, and a better understanding of such short-lived, low-precipitation, and high-impact coldseason convection lines.Following this introduction, Section 2 reviews the event with station observations and large-scale circulation data,Section 3 shows the role of frontogenesis, Section 4 diagnoses the vertical motion with the Q -vector equation, and Section 5 provides a summary and outlook for future work.

2.Surface observations and synoptic circulation

The surface observations are shown in Fig.1 (a).The total snow accumulation in 12 h ending 1200 UTC 26 January demonstrated a banded structure, extending from Liaodong Bay to Heilongjiang Province.The amount at most weather stations was less than 5 mm.Fig.1 (b) gives the time series of surface observations at Shenyang.The main features are as follows: (1) the dewpoint deficit decreased rapidly from 0300 UTC, supporting a near-saturated environment for the snow burst, and the cold front was substantiated by an obvious decrease in temperature and dewpoint from 0500 UTC; (2) the hourly precipitation concentrated from 0500 UTC to 0800 UTC and the total was small; and (3) a wind shift from southerly to westerly or northwesterly was observed around 0500 UTC, corresponding to the report of snow and passage of the front.The surface observations suggest this was a brief but intense burst of snow.

The fifth major global reanalysis produced by ECMWF (ERA5), with a horizontal resolution of 0.25° and vertical resolution of 37 pressure levels, was used to analyze the upper-air situation.At 0600 UTC 26 January 2017, the 200-hPa tropospheric flow ( Fig.1 (c)) showed a zonal jet extending from central Inner Mongolia to Bohai Bay.Northeast China was located on the left side of jet exit region, where conditions are typically favorable for ascent.The 500-hPa synoptic-scale flow was characterized by cyclonic flow across Northeast China with a trough, enhancing the surface cyclone and producing ascending motion in the frontal trough region.At 850 hPa, a cold front was evident, as indicated by the strong equivalent potential temperature gradient, and was associated with a wind shear line extending in the northeast–southwest direction.The shear line gave rise to local convergence and lifting, which released unstable energy and stimulated the convection line.

The above analysis shows that the coupling of the high-level jet, midlevel trough, and low-level front provided favorable large-scale conditions for ascending motion over Northeast China.

3.Analysis of frontogenesis

Thermal direct circulation driven by the frontogenesis ahead of a cold front is believed to be the lifting mechanism that triggers instability( Markowski and Richardson, 2010 ).A 2D version of frontogenesis that more clearly reveals the regions with thermal circulation is given by( Martin, 1998 ):

The horizontal distributions of frontogenesis in Fig.2 were examined.The convection, indicated by vertical integration of hydrometeors,consisted of two parts at 0200 UTC ( Fig.2 (b)): a narrow convection line ahead of the cold front, and stratiform cloud over Heilongjiang.Weak convections were also found over the high terrain (Changbai Mountains)(41°N, 124°E) and remained quasi-stationary.There were two separated maximal frontogeneses present in the low troposphere.One was coincident with the cold front and convection, serving as the primary forcing to release unstable energy, and the other was oriented parallel to the Changbai Mountains on the windward slope.As the frontogenesis moved downstream, the convection line was lengthened, strengthened,and intersected with the mountains.By 0600 UTC ( Fig.2 (c)), the northern frontogenesis merged with the frontogenesis on the western slope of the Changbai Mountains, with a significant increase in strength of the convection line.The wind shifted on the two sides of the frontogenesis,which was in favor of airflow convergence.Subsequently, the frontogenesis moved up against the mountains and further enhanced ( Fig.2 (d)),indicating the combined role of the frontogenesis and high terrain in the development of this snowband.

The vertical structures along the black line in Fig.2 (a) are displayed in Fig.3.At 0000 UTC, the frontogenesis coincident with the cold front was present near the surface at (43.3°N, 121.2°E) and tilted westwards with height.It followed the isentropes and was associated with intense convergence.The airflow divergence occurred right above 900 hPa ( Fig.3 (a)), constituting the typical pattern favorable for vertical motion.A low-level jet ( Fig.3 (a)) was located at 850 hPa near(42.2°N, 122.5°E), transporting warm and moist air and producing remarkable convergence on its left-hand side ( Fig.2 in Jiao et al.(2020) ).The equivalent saturation potential temperature ( Fig.3 (b)) decreased from 850 hPa to 700 hPa at (43.4°N, 121.3°E), indicating a region of conditional instability (CI).The coupling of low-level frontogenesis and convergence caused strong ascending motion, contributing to the release of instability.The frontogenesis around the Changbai Mountains was located downstream within convergence regions.One band was located at the foothills near (41.8°N, 123.6°E), tilting forwards with height,and the other was located over high terrain from (41.0°N, 124.8°E) to(39.8°N, 126.4°E).There was a clockwise circulation on the windward slope.Near-surface wind moved downwards along the windward slope in the divergence region, then rose along the frontogenesis, and finally flowed backwards near 850 hPa.By 0600 UTC ( Fig.3 (c,d)), the regions of CI near the surface and limited regions of pure conditional symmetric instability (CSI) and inertial instability (II) appeared at the foothills,near (41.8°N, 123.6°E), within the convection, which was also found in Jiao et al.(2020).The merged frontogenesis ( Fig.2 (c)) arrived at the foothills, resulting in the significant convergence owing to the combined effect with the mountains.The pattern above the terrain did not change much.However, the intensity of frontogenesis and convergence increased obviously, as well as the intensity of the clockwise circulation whose northern updraft merged into the slanted updraft.The jet moved over the mountains and the ascending branch on its left-hand side overlapped with the frontogenesis band at the foothills.All of these factors contributed to an intense updraft triggering instabilities and enhancing the snowband.

Fig.5.Schematic diagram of (a) the process leading to the formation of the snowband far away from the high terrain, and (b) the process leading to the enhancement of the snowband under the influence of high terrain.See text for detailed discussion.

The above analysis suggests the trigger during the initial stage was mainly caused by the frontogenesis and low-level jet.During the enhancement stage, the trigger was a combination of the frontogenesis,local terrain circulation, and low-level jet.

4.Diagnosis of vertical motion

The

Q

-vector is usually used to diagnose vertical motion.Ran et al.(2019) proposed a modified

Q

-vector suitable for heavy rain by introducing the generalized potential temperature, which is sensitive to water vapor variations and closely related to precipitation.The Omega equation is given by:

The vertical structure of Q -vector divergence and its components are depicted in Fig.4.At 0000 UTC (left-hand column), the negative Q -vector divergence is remarkable at A, B, and C in Fig.4 (a).C extends from 900 hPa to 750 hPa and follows the moist isentropes, causing vertical ascent and promoting convection.The negative Q -vector divergence at A and B results in the upright ascent near the surface and slanted ascent above 900 hPa between (42.5°N, 122.4°E) and (41.8°N, 123.6°E).This contributes the release of instability and snowband development in the following hours.Near the Changbai Mountains, downstream, the Q -vector divergence is negative at the foothills and over the high terrain near 900 hPa, associated with the local vertical motion.The pseudovorticity term ( Fig.4 (c)) dominates at B, suggesting a significant role of wind shear associated with the low-level jet.The frontogenesis term( Fig.4 (e)) dominates at A and C, indicating the contribution of low-level frontogenesis to vertical motion.Both played an important role in the ascending motion during the initial stage.As the convection moves to the foothills at 0600 UTC, the negative-value regions of Q -vector divergence are located at the western slopes of the Changbai Mountains near(41.8°N, 123.6°E), with significant enhancement.They are restricted within the convection and tilted forwards with height.This pattern is accompanied by strong and narrow ascending motion, extending up to 700 hPa.Among them is a significant positive region promoting descending motion, which is not captured in the ERA5 dataset, possibly because of its limited resolution.The vertical structure of frontogenesis term ( Fig.4 (f)) is almost the same as that of Q -vector divergence, which suggests frontogenesis was the main mechanism for ascent at this moment.It is also worth noting that the pseudo-vorticity term ( Fig.4 (d))is intense over the high terrain, especially near (41°N, 124.8°E), due to the wind shear associated with the low-level jet and terrain effect.This proves that the low-level jet and terrain played an important role in the ascending motion.

Finally, it is interesting to compare the frontogenesis vector divergence term in the Q -vector with the 2D frontogenesis in Section 3.At 0000 UTC, the regions with frontogenesis vector convergence ( Fig.4 (e),A and C) are generally in agreement with the maximal low-level frontogenesis in Fig.3 (a).However, there are significant differences at 0600 UTC.Firstly, the frontogenesis near (41.8°N, 123.6°E) in Fig.3 (c) extends to 850 hPa, while the corresponding frontogenesis vector convergence (A) in Fig.4 (f) extends higher, to 750 hPa.Secondly, the region of frontogenesis vector convergence (C) in Fig.4 (f) is upright and located within the convection.However, the corresponding frontogenesis band in Fig.3 (c) is tilted westwards following the dry isentropes.The difference is attributable to the uneven distribution of water vapor, which further substantiates the close relationship of the Q -vector including generalized potential temperature with precipitation.

The above analysis indicates that the wind shear and frontogenesis combined to result in Q -vector convergence and ascent during the initial stage.The frontogenesis effect dominated when the convection approached the windward slope of the Changbai Mountains.It is also proven that the Q -vector divergence accounted for the vertical motion.This means Q -vector divergence can serve as a potential tracking method for snowbands.

5.Discussion and conclusion

During the daytime of 26 January 2017, a linear snow burst event moved across Northeast China, producing brief but heavy snow.The snowband was generally aligned with the surface cold front and oriented in the northeast–southwest direction, parallel to the direction of the Changbai Mountains.The ERA5 dataset was used to carry out a preliminary analysis.The convection firstly occurred in the middle troposphere, associated with low-level frontogenesis and negative Q -vector divergence.As the band approached the Changbai Mountains, it was influenced by the terrain and experienced a significant enhancement.

The frontogenesis and modified Q -vector was analyzed to diagnose the possible trigger mechanism during the initial stage and enhancement stage.The results are summarized as a conceptual model presented in Fig.5.During the initial stage ( Fig.5 (a)), the low-level southwesterly jet transported warm and moist air and produced remarkable convergence on its left-hand side.Surface frontogenesis and convergence at the leading edge of the cold front, corresponding to the negative Q -vector divergence region, resulted in ascent that released instability, contributing to the initialization of convection.When the band was located at the foothills of the Changbai Mountains ( Fig.5 (b)), a combination of CI, II, and CSI was released by the horizontally restricted ascending motion caused by the low-level frontogenesis, terrain circulation, and strong wind shear associated with the low-level jet.Negative Q -vector divergence regions coincided with the convection and ascending motion zone.

Some outstanding problems still remain: (1) A detailed picture of the mesoscale characteristics has not yet been established, especially the positive Q -vector divergence present upstream of A in Fig.4 (b); (2) the effect of latent heating on the ascending motion needs further investigation; and (3) the extent to which the terrain affects the frontal development and snowband enhancement requires further study.In follow-up work, a high-resolution simulation and several sensitivity tests will be carried out to obtain a more detailed result.

Funding

This study was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences [grant numbers XDA17010105 and XDA20100304 ], the National Key Research and Development Program [grant numbers 2018YFC1507104 and 2019YFC1510400 ], the Key Projects of Jilin Province Science and Technology Development Plan [grant numbers 20180201035SF ], and the National Natural Science Foundation of China [grant numbers 41775140 and 41790471 ].