NHESSDNatural Hazards and Earth System Sciences DiscussionsNHESSDNat. Hazards Earth Syst. Sci. Discuss.2195-9269Copernicus GmbHGöttingen, Germany10.5194/nhessd-3-3449-2015Review Article: Explosive cyclogenesis over the south-east of Romania 2–3 December 2012BratuM.marius.m.bratu@gmail.comNichitaC.National Meteorological Administration, Timisoara, RomaniaM. Bratu (marius.m.bratu@gmail.com)22May2015353449348524February20157April2015This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://nhess.copernicus.org/preprints/3/3449/2015/nhessd-3-3449-2015.htmlThe full text article is available as a PDF file from https://nhess.copernicus.org/preprints/3/3449/2015/nhessd-3-3449-2015.pdf
This paper is devoted to the study of the synoptic-dynamical conditions that
contributed to the development of a rare explosive cyclogenesis event that
occurred at the beginning of the winter from 2012 to 2013 in south-eastern
Romania, more precisely between 2 and 3 December 2012. The minimum sea level
pressure observed was 980.2 hPa, the lowest ever observed record for
the surface of the Sulina weather station, and also over the western side of
the Black Sea during the of period 1961–2000 and 1965–2004. It was found
that the cyclone was not a regular one, but a real “meteorological bomb”
one, where the central pressure at sea level recorded an extraordinary
decrease at about 32.3 hPa in 24 h, equivalent with 1.7 B (Bergeron
unit). Compared to the 20th century storms named Lothar and Martin (level 2
and 1 on the hurricane scale) which devastated western and central Europe in
December 1999, this case of explosive cyclogenesis can be considered one of
the most extreme for our area, from both a meteorological view as well as its
effects.
Introduction
There are quite frequent situations in some areas as the
Atlantic
Ocean, the Pacific Ocean and the east coast of the USA (Roebber, 1984;
Sanders, 1986; Gyakum et al., 1989), often also behind the seas with
important surface thermic gradients of the water, where the
transformation of a cyclone from the wave status to the maturity
status happens so fast and with such intensity, that the surface
atmospherical pressure reaches critical values in a short time, being
characterized by abundant rains, strong winds (average wind >17ms-1) and dangerous significant wave heights (swell)
for human activities, such as sea and air navigations. By definition,
the extra tropical cyclone settled to a precise latitude, whose
pressure decreases in its centre with a ratio of 1 hPah-1
during 24 h (Sanders and Gyakum, 1980) is called generally
“meteorological bomb”.
An analysis of meteorological conditions, such as the heat flux,
moisture budget and upper-air features was performed by Gyakum and
Danielson (2000) and Strahl and Smith (2001). Studying a large number
of weak and “bomb” events over the Pacific, they found out that
large surface fluxes are crucial in the case of explosive
cyclogenesis. Recently, Nielsen and Sass (2003) in their study of the
North Sea severe storms have also identified the precursors, such as
potential vorticity (PV) generated by latent heat release (LHR) that
brings a major contribution to the deepening of the studied
cyclone. Some important studies were done by Lagouvardos (2006) and
Michele Conte (1986) closer to our geographical area, in the
Mediterranean Sea, where the most frequent cyclogenesis processes that
influence Romania's weather-climatic environment happen, especially
during the cold season.
This work is dedicated to the investigation of a rare explosive
cyclogenesis event that happened over the south-east of Romania and
the western side of the Black Sea Basin, when an absolute record for
the lowest sea level pressure (980.2 hPa) was recorded at
Sulina meteorological station (15360) in December 2012. This value was
37 hPa lower than the previous one observed 23 h
before. Sustained winds and gale winds in eastern Romania and the
western part of the Black Sea exceeded 20–25 ms-1 on
average, with gusts up to 38 ms-1, which resulted in all
the harbours being closed, the trees blown down and the roofs
dislocated and damaged. The power supply network was interrupted and
the eolian power farm networks were out of order. The gradual filling
of the cyclone slowly evolved in the next 15 h (with only
4 hPa pressure difference from the lowest one) until it
reached the mature stage. On the back side of the low centre and near
it, harsh weather conditions swept across the south-eastern Romania,
the Pontic Coast and the western part of the Black Sea, also with
important 12–24 h of total rain accumulations (torrential rain
between 20 and 60 Lm-2, with peaks at about
69…70 Lm-2) and blizzards in the mountainous area
– Oriental Carpathian Mountains.
The “bomb” analysed in this work had a Mediterranean–Aegean origin
and its evolution occurred on an unusual Trans-Balkan trajectory
(Fig. 1) and deflected towards the North, in comparison with the
classic one (type 2b'), as it was settled by Sorodoc (1962) and
revised by Ecaterina Ion-Bordei (2009). It was concluded that the
rapid deepening of the cyclone was associated with a short and rapid
trough system (Stratospheric dry air intrusion, Fig. 4a) that, upon
the influence of a very intense subtropical upper-level jet (STJ),
merged into a steady baroclinic low-tropospheric environment still
existent on the sea coast of Romania and Bulgaria. Prior to the rapid
cyclogenesis, the growing trough acquired a negative tilt which
intensified the process (Gaza and Bosard, 1990). Based on WRF-ARW
non-hydrostatic mezzo-scale limited area model simulations (no coupled
model developed in this study, CPLD), we could conclude that the
upper-level dynamic factor (trigger) was not the main reason that led
to the growth and decay of the explosive cyclogenesis under study. The
significant level of the surface latent heat flux (diabatic heat)
released within the cyclone, during its explosive phase, that
generates an intense low-level vortex (in terms of potential vorticity
values observed at about 3 to 5 PVU for at least 9 h), contributed
as a crucial factor to its rapid deepening. The most recent studies
show that a subsequent intensification could be the result of the
occurrence and growth of a diabatic Rossby wave (DRW) and of the
nonlinear interaction of its short wave (mezzo-scale 200 km)
with an upper-level disturbance (Moore and Montgomery, 2004). Section 2 is
devoted to the climatology description of the
phenomenon through the investigations of ECMWF analyses, historical
observations and different authors. An insight into the method and
data used in this work is given in Sect. 3, which is based on the
Bergeron unit calculus. Section 4 describes the synoptic environment
under which the deepening of this cyclone occurred, and also helps to
understand its damaging wind surface intensifications. The PV and DRW
analyses based on WRF-ARW model output are shown in Sect. 5 followed
by the storm weather impacts, while concluding remarks are provided in
the final section.
This is the first approach to the synoptic-dynamic setting of this
type of cyclogenesis event over the Romanian territory.
Phenomenon climatology
Sanders and Gyakum (1980) presented the first climatological study of
this subject when they introduced the notion of explosive cyclone
development or “bomb” cyclogenesis. But their study was focused over
the Pacific and North Atlantic regions. In time, many other authors
have come with additional explanations and interpretations. In the
past, few investigations were done in the Mediterranean Basin, but
recently some Greek and Italian authors such as Lagouvardos (2006) and
Brunetti (2005) have focused on this area. The Mediterranean Sea, due
to its complex and particular geographical configuration as a closed
and warm sea, is the reason why these phenomena are relatively
frequent especially during the winter season. Because this phenomenon
is very rare at the latitude of Romania and there are not sufficient
studies regarding the Black Sea region, the majority of cyclogenesis
having Mediterranean origins, we have considered it useful to have the
studies made by Brunetti and Moretti (2005) in the entire
Mediterranean Sea Basin as important reference.
From 1980 to 2004, 79 such events occured over the Mediterranean Sea,
an average of 3 year-1. The meteorological “bombs”
started in October and ended in May. The maximum number of events
happened in December (18 events) when the sea is generally still very
warm, followed by January with 15 events while the minimum of events
happened in May. Figure 2 indicates the frequency of the critical
values expressed in Bergeron units (B), values observed between 1980 and
2004. The biggest drop of 1.59 B was reached only once (on 21 January
1981), when the pressure reached 27 hPa in 24 h at
the latitude of 39∘ N. Similar events have never been
recorded in June–September. Brunetti and Moretti (2005) also provided
a very useful list of the meteorological bombs from the Mediterranean
Sea, which contains: the date, the initial and final pressure values
and also the differences between them, the recorded Bergeron value,
the latitude and longitude of the maximum deepening, the geographical
area and the type of cyclogenesis (e.g. frontal, continental,
African).
An important cyclogenesis is the Aegean source type (Fig. 1), with 6
events in a quarter of a century. In our case the cyclogenesis was
initiated also in the Aegean Sea Basin, and in less than 24 h the
central pressure of 32.3 hPa measured at 12:00 UTC on
2 December fell to 06:00 UTC on 3 December (Fig. 3, left)
corresponding to 1.7 B and thus the low pressure system can be
referred to as a strong “bomb”.
According to the MEDEX (MEDiterranean EXperiment), the database
constructed by ECMWF reanalyses the model, this case being the deepest
cyclone observed in the entire Romanian area, including the western
part of the Black Sea Basin between 1965–2004. At that time the
Sulina weather station reported an extraordinary pressure fall of
37.5 hPa in 24 h (Fig. 3, right) and a minimum station
pressure of 978.9 hPa; the minimum can be considered a record
after the latest record value of 980.9 hPa (Table 1) according
to 1961–2000 NMA database records.
There are a lot of studies starting with Sanders and Gyacum (1980)
that describe the synoptic upper air aspects of explosive
cyclogenesis. The main factors are: horizontal temperature gradient,
surface heat fluxes, diabatic heating, air–sea instability, jet
interactions and tropopause folds. Examining a number of cases, Bruce
and Elmar (1988) made an analytical study of the continental bombs
development over the Eastern USA, making comparisons between a regular
cyclogenesis and an explosive one. They found significant signatures
especially in divergence, vorticity advection and pressure tendency,
latent heating and static stability which allow distinctions between
bombs and regular cyclones. More recently, based on Parker and Thorpe (1995),
Moore and Montgomery (2004), upon reexamining the dynamics of
short-scale, also established that the diabatic Rossby waves growth
mechanism may play an important role, considering it a precursor for
an explosive cyclogenesis.
Methods and data
The method used in this work was discovered by Tor Bergeron
(1891–1977), who studied the synoptic climatology and the motion of
this process of the extra-tropical cyclones in the North Atlantic
Ocean for the first time. He established the following principle to
identify a meteorological bomb that has been used until today: the
pressure in the centre of a cyclone must decrease by at least
1 hPah-1 during a period of 24 h. The Bergeron
definition used by Sanders and Gyakum (1980) to analyse the concept of
explosive cyclogenesis (“bomb”), had as reference the 60∘ N
latitude. In order to monitor the intensity and track of this type of
cyclone step by step, the Bergeron formula was used:
NDRc=Δpc24⋅sin60∘|sinϕ∘|
where Δpc represents the pressure variation in the
center and ϕ represents the latitude. The Bergeron formula allows
us to obtain a critical NDRc ratio (normalized
deepening rate of central pressure) according to the latitude we want
to use. When the pressure variation is of 1 hPah-1 and
ϕ=60∘, the NDRc=1, namely the
threshold value of 1 B where the explosive cyclogenesis can start to happen:
1hPah-1⋅sin60∘/sin70∘=0.92hPah-1
The measurement obtained by Bergeron indicates the decreasing speed of
the atmosphere pressure in the depression center settled to a certain
latitude. The critical ratio of 1 B is considered a reference value
for a meteorological bomb. This indicator can vary from 28 hPa
in 24 h (at the poles) to approx. 9 hPa in 24 h (at
20∘ N lat), this last value being calculated at the southern
limit where the phenomenon has been observed until now. The Italian
National Weather Service from the Mediterranean Sea Basin has adopted
a critical value of 1 B that represents a lowering of the
pressure in the center of at least 17 hPa in 24 h (which
corresponds to an average latitude of 38∘ N). In the case of the
Black Sea Basin, we have considered 19 hPa as the critical value for
1 B (19.6 =24/1.22;
1 hPah-1⋅sin60∘/sin45∘= 1.22 hPah-1), for the medium value of
45∘ N. In our case, the lowest value of the sea level
pressure was 980.2 hPa (37.5 hPa in 24 h) recorded in
Sulina (Dobruja), and the highest value of the slp was
1012.5 hPa registered in the initial stage of the cyclone over
the Aegean Sea. Considering as reference the pressure variation of
32.3 hPa (1012.5-980.2=32.3) in 24 h in the center of the
low-pressure, in the case of our country the critical ratio has reached the
extreme value of 1.7 B (32.3:19=1.7),
in comparison to our neighbour (the Mediterranean Sea basin), where
the value of 1.59 was rarely reached (Brunetti and Moretti, 2005).
In this way the indicator NDRc was monitored for
15 h (every 3 h) and the conclusion is that there is a gradual
increase of its value from 1 to 1.7 B, from the explosive
deepest status of the cyclone (Fig. 4a–d) up to the maturity status
(Fig. 4e and f). The main meteorological data and sources used in this
paper are presented below:
Daily SYNOP and METAR messages mainly at 00:00, 03:00,
06:00, 09:00, 12:00, 15:00, 18:00, 21:00 UTC used for issuing 24 h
weather diagnosis (T 2m, Td, mixing ratio, MSLP, pressure tendency,
wind direction and speed, 24 h total precipitation, sea surface
temperature, high sea wave) at all the met stations (RMS) and also
radar observations belong to the Romanian Meteorological
Administration NMA network; for the worldwide data we also used daily
SYNOP and METAR messages provided via internet by Wyoming, Albany and Florida
State Universities.
Daily SOUNDINGS messages at 00:00 and 12:00 UTC
2/3 December provided via internet by Wyoming, Albany and Florida
State Universities.
Six hourly analysis ECMWF model at surface and upper
levels (6 pressure levels at SLP, 1000, 850, 700, 500 and
300 hPa) different parameters (temp., temp. advection,
geopotential, divergence, potential vorticity in pressure level hight,
streamlines, max. wind), soundings and cross-section provided by
EUMETNET.
Daily analysis SKIRON model (horizontal resolution
∼5km) about the Sea Surface Temperature and SST
anomaly, sea surface height above the sea level provided by
www.myocean.eu/.
Hourly analysis and forecast WRF-ARW 3.4.1
(4.5 km hor. resolution using 00:00 UTC GFS reanalysis on
a Linux HPC Cluster Infragrid Cluster) non-hydrostatic limited area
model surface mean-sea-level pressure, surface latent heat, potential
vorticity (PVU unit) at 925, 850 and 887 hPa levels made by
RoMetEx and West University of Timişoara, Romania.
Hourly and six-hourly geo-stationary and polar satellite pictures (RGB,
HRV, IR, WV) provided by Eumetsat and University of Dundee, UK.
MEDEX (MEDiterranean Experiment) database (constructed
by ECMWF analyses and ERA-40 reanalyses), available on
http://medex.aemet.uib.es.
Climatological database from NMA Atlas, Romania.
The hourly (for the surface) and 12 hourly (for the upper level)
varied analyses have been computed using Digital Atmosphere Soft V2.07
(Weather Graphics) on the meso-α and meso-β scale. For
this purpose we have used objective analysis methods such as the
Nearest Neighbor and the Cressman type.
Synoptic overview
This event took place in quite abnormal climatic conditions, with
a neutral NAO index and a slightly better negative AO index
(-1…-1.5). A low pressure area dominating the central
Mediterranean Basin area was linked to a large persistent upper Rossby
wave (a geopotential negative anomaly), that remained in Western
Europe for 3 days (Fig. 5). A maritime arctic airmass came from
north-west Europe and reached Italy and the Balkan Peninsula. The main
cold front separated the chilly unstable Italian air mass from the
warm, also unstable eastern Mediterranean air mass. Also, both the
Mediterranean and the Black Sea experienced positive SST anomalies in
December. Regarding the Black Sea surface temperature, it was between 11 and
14 ∘C, 3–6 ∘C higher than the monthly normal
values.
Initiation phase I
At 00:00 UTC on 2 December, the cyclone being analyzed here had not
been formed yet. At 12:00 UTC on 2 December, a low-pressure area in
the incipient phase with 1008.3 hPa in its center was located
in the maritime eastern area of Greece (Fig. 6a). There were two
disturbances, a short-wave over the area of the Ionian Sea with its
axis extending NW–SE (negatively tilted) and a cut-off low over the
Sicilian Channel, observed at the 500 hPa constant pressure
level (Fig. 7a and b). A strong upper-level subtropical jet (STJ)
accompanied the cut-off wave, reaching a maximum value of ∼45ms-1 at 300 hPa (Fig. 7b).
Initiation phase II
During the following 12 h, at 21:00 UTC on 2 December, the low
pressure was in its incipient phase II in the southern continental
Bulgaria with a central pressure of 998.8 hPa
(Fig. 6b). Regarding the 500 hPa two-trough system
(∼ 5500 ma.s.l.), we can observe geopotential troughs in the
western part (over the Sicilian Channel), almost quasi-stationary
inoculated, moving on a SW–NE trajectory under the influence of the
Subtropical Jet. At the 300 hPa level an intensification to
56–60 ms-1 over the southern edge of the trough can be
observed and also a detachment over the northern edge of the trough of
an isolated upper level jet in diffluent flow (Drift), accompanying
the cyclogenesis process (Figs. 7c and 9a). At the same
500 hPa level, as a band, we can see two positive vorticiy
advection areas (Fig. 8b and c): one over the east of the Aegean Sea
and the Marmara Sea towards the low centre, and another one, less
intense, over the eastern Romania, Moldavian Republic and southern
Ukraine. The trough axis is still negatively tilted, thus generating
cyclonic vorticity (Gaza and Bosard, 1990). It can be noted that the
cyclonic vorticity advection as a dynamical factor begins 12 h before
the cyclone development (Fig. 8b) and it is more obvious over eastern
Romania and the western part of the Black Sea Basin. Warm advection at
the same level is also obvious over the Black Sea, Moldavian Republic
and Ukraine (brown bold line, Fig. 8b and c). In the following 9 h on
3 December, the low pressure system moved northwards, rapidly
deepening first to 991.2 hPa at 00:00 UTC and then to
985.0 hPa at 03:00 UTC (Fig. 6c). The explosive phase had
just started. The intrusion of stratospheric dry stable air in the
middle and lower troposphere levels, visible on the satellite infrared
WV Channel as a black tongue shape (Santurette and Georgiev, 2005)
indicates the presence of an active dynamic tropopause anomaly (DTA).
It was identified based on the analysis of the Ertel's potential
vorticity (PV) field (magenta contour lines, taking 1 PVU from the
dynamic tropopause as a reference value, Fig. 9a and b). The maximum
potential vorticity area (PV max) became gradually narrower. The
intrusion of stratospheric dry air created upward motions and local
instability that led to the organization of a deep convection into
a line on its eastern flank. At 06:00 UTC on 3 December, the
mezzo-scale low-pressure reached its minimum central pressure
(980.2 hPa) over the northern Dobruja (Fig. 7c). According to
Sanders and Gyakum (1980), explosive cyclogenesis occurs when the
deepening of the cyclone exceeds 1 B. In the area over the
Black Sea (at ∼45∘ N lat), 1 B equals
19 hPa in 24 h. Therefore the 24 h central pressure fall
from 1012.5 hPa at 12:00 UTC on 2 December to 980.2 at
06:00 UTC on 3 December corresponds to 1.7 B and thus to
a strong “bomb”. Important modifications of the dynamic nature
occured in the upper-level above the 500 hPa level, 6 h
before the beginning of the explosive phase. The curvature trough
amplified and the gradient of the geopotential height
deepened. Meanwhile, in the low-levels, a critical potential vorticity
anomaly at about 4.5–5 PVU appeared in a comma shape, being
discussed in detail in Sect. 5. As inferred by the strong pressure
gradient (∼13hPa pressure change in just 100 km
range around the storm centre), the prevailing damaging winds, first
in the eastern sector and then in the western sector, reached their
maximum intensity in south-eastern Romania, over the Black Sea coast
and off shore between 2 and 3 December. The local weather stations
registered high intensity widespread winds exceeding
20–25 ms-1 and a gust of over
32 ms-1. Maximum wind gusts at Gloria Oil Platform
(15477), Medgidia (15462) and Sulina (15360) reached
35–38 ms-1, which means 137 kmh-1, the
first level on the Saffir–Simpson hurricane scale, the same as for
the Lothar and Martin cyclones that devastated western and central
Europe in December 1999. The strong gusts which were recorded appear
to have been due mainly to the associated sting jet (SJ) in the low
levels by formation of a bent-back front (BBF; Shapiro and Keyser,
1990; Neimann et al., 1993). According to the Browning study (2004),
the SJ produces strong winds that reach the surface in the dry air
just ahead of the tip of the hook-shaped cloud head that accompanies
the BBF. However it is better not to lose sight of the fact that the
strength of a cyclone is first determined by the interaction of the
over-running potential vorticity anomaly with the baroclinic zone, and
the SJ is only a secondary factor in creating a local intensification
of the wind. Based on the Medgidia radar Doppler velocity field and
the numerical model transverse cross section, we concluded that the SJ
was responsible for stronger surface wind intensifications, but only
for a short period. The Fig. 11b and c (right) indicates well the SJ
location in its ascending (south-western) and in its descending
(north-western) trajectory during the explosive cyclone phase. The
satellite images provide a clear overview to describe the mezzo-scale
evolution of our storm. Shortly after an explosive stage, a clearly
defined “eye” along with spiral structured cloud bands were visible
from the MSG-2 satellite RGB images on 3 December at 07:00 UTC
(Fig. 12). Also, the polar NOAA-MODIS images show at a higher
resolution the location of the low in the neighborhood of Odessa City
(Ukraine) with a clearly defined “eye” (Figs. 12 and 13).
Mature phase
In the next 12 h the cyclone reached a mature stage moving towards
north-east. The cyclonic perturbation is visible at all levels both at
the surface and in the upper-air (Figs. 6d, 7d and 8d). On the
4 December, the cyclone dissipated over the northern part of the
Russian Plain, after moving over ∼2000km at an average
speed of 50–55 kmh-1 in the explosive phase.
Results
During the explosive phase, the largest deepening occurs from surface
to 800–850 hPa and it is related to an intense moisture
convergence field coming from the Black Sea (the eastern sector)
towards the cyclone center (Figs. 14a, 15a and 16a) within a significant
positive SST anomaly up to 5–6 ∘C (14b, 15b and 16b). This
is connected to an intense low-level vortex associated
with an ascending moist air movement that produced a rapid pressure
drop, condensation, latent heat release and a thermal profile that is
specific for warm cyclones beetwen 850 and 700 hPa (not
shown). A closer analysis of the potential vorticity anomaly field
provided us with the idea that the subsequent intensification could be
the result of a growing diabatic Rossby wave (DRW) and of the
nonlinear interaction of its short wave (small-scale less then
500 km) with the upper-level disturbance (Moore and Montgomery,
2004). The WRF-ARW 3.4.1 non-hydrostatic numerical model
was used in a one nested grid configuration, with a finest horizontal
resolution of 4.5 km, to simulate the genesis and the
evolution of the low-level vortex (potential vorticity) and also the
surface latent heat flux. The simulations were initialized with GFS
model reanalyses input at 00:00 UTC on 3 December 2012. Figures 14c, d,
15c, d and 16c, d show a series of potential
vorticity maps (between 925 and 850 hPa) at 01:00, 03:00 and
06:00 UTC on 3 December 2012. More specifically, the Ertel potential
vorticity (defined as q=1ρη∇θ, where
ρ is the density, η is the absolute vorticity vector and
θ is the potential temperature) is given in PVU (potential
vorticity units: 10-6km2kg-1s-1). A few hours
prior to the explosive phase, around 00:00 UTC on 3 December, near the
border line between Romania and Bulgaria (Fig. 14c and d), the high PV values
at about 3.5–4 PVU appeared first at the lower level of 925 hPa
(600–700 m high), while at 850 hPa level
(1300–1500 m high) the values were of only 2 PVU. In the
next three hours, over the continental part of Dobruja district, the
PV magnitude reached the phenomenal 5 PVU on both levels
(Fig. 15c and d). Note that the maximum PV (red shaded contours)
coincides very well with the maximum moisture convergence area and
stream current confluence lines. This means that the generation of
diabatic wave (diabatic Rossby wave) appears to be tied with the
moisture convergence field and with the maximum low-level jets
intensity in the south-east sector (not shown). In the mature phase,
the low-level vortex was dispersed in the upper troposphere by the
presence of cyclonic vorticity advection. The model maps highlighted
that the PV magnitude was maintained at high values as the same
previous values (4–5 PVU) for another 5–6 h while the low-pressure
system was away from the Black Sea area. In the next hours, the
diabatic wave started to rotate into the seclusion front
(Fig. 16a–d), reaching the minimum historical surface pressure of
978.9 hPa. The role of the surface heat flux has been
investigated in many cases of explosive (as well as ordinary)
cyclogenesis and it is considered to be a crucial feature during the
cyclogenetical process (Gyakun and Danielson, 2000). Since 3 December,
03:00 UTC, the heating flux (just the surface latent heat shown)
gradually increased during the explosive phase reaching a great
evaporation rate at about 330 Wattm-2 from 07:00 to
10:00 UTC (Fig. 17). For comparison, a “hurricane-like” system
could easily gain up to 600–800 Wattm-2. The presence of
a quasi-concentric surface heat distribution around the storm eyewall
is noteworthy. In addition, it should be emphasized that the
reasonable parametrization of the atmospheric-ocean-wave interaction
in the numerical model is quite important and essential for a reliable
extra-tropical cyclone intensity and accurate central pressure
forecast.
Weather impacts
The phenomena associated with this “meteorological bomb” consisted
not so much in the accumulation of precipitation, but especially in
wind gusts that reached level 1 on the Saffir–Simpson hurricane scale
(85 mph or 137 kmh-1). Also, off the Black Sea coast the
waves reached 10 m in height and the water retracted
40 m towards the coast, also recording very strong rip
tides. At that time, the local and national mass media related the
damage of this explosive cyclogenesis. The disastrous effects in
Constanta and Tulcea counties during the night of 2–3 December 2012
were: the winds reached over 100 kmh-1 and tens of trees
and electricity poles were blown down, national roads were blocked and
electricity was interrupted for many hours. Only in Constanta county
28 cities remained without water, the gas network was damaged, 22
schools and 348 houses were damaged. In the case of the schools, the
financial loss was around RON 1 888 000. In Medgidia, the roof of
a house was blown away and a high school in Poarta Alba also remained
without the roof, all the windows being broken and some walls
dislocated. The 8 level storm (on the Douglas scale) which started on
the sea forced the local authorities to close the ports at the Black
Sea. In Galati the strong wind blew away the advertising panels, the
trees and the road signs. In Braila, a tree blocked the tram lines and
the storm was followed by a rainfall recording
40 Lm-2. In all Dobruja area, half of the eolian stations
were out of order for at least 24 h and the nuclear reactor from
Cernavoda was turned off for technical verifications only 3 days
after this event finished. The most quantitative rains were recorded
in Slobozia and Ialomita counties; in less than 12 h the water
quantities recorded values between 30 and 69 Lm-2,
resulting in a flood. In Slobozia a peak of 70 Lm-2 was
recorded in less than 12 h. Important floods were also recorded in
Barlad county. Ukraine and the Republic of Moldavia were also
affected. The strong winds and rainfalls interrupted the electricity
in 80 villages, blew away the roofs, and in Chisinau many streets were
flooded.
Conclusions
The analysed case fits well into the criteria established by
Sanders
and Gyakum regarding the explosive cyclogenesis, as a predominantly
maritime type (inland during the explosive phase), specific to the
cold season, in the maximum deepening phase having the features of
a tropical cyclone (hurricane), both in the wind field (powerful wind,
devastating swell) and in terms of cloud appearance (spiral shaped
cloud system at sub-synoptic scale, eye formed above the cyclone
recorded on satellite images). The superficial surface temperature of
the water in the western and southern basin of the Black Sea showed
a significant discontinuity and a positive anomaly of
3–6 ∘C, which, on an eastern circulation, increased
considerably by the convergence of warm and humid air contribution to
the centre of the cyclone. Given the rarity of this phenomenon in our
geographical region, unprecedented in recent history from the strength
magnitude point of view, it is necessary to reconsider this type of
rapid cyclogenesis. In this research, which is not an analytical one,
we have tried to investigate which factors (diabatic or dynamic) have
the largest contribution in its evolution. Although the cyclogenesis
was initiated over the Aegean Sea by classical processes such as the
positive advection of vorticity and the divergence in the upper
troposphere, we have found that the deepening of the “bomb” in
south-eastern Romania took place under the diabatic mechanisms rapidly
associated to these processes, that generated for a short time
(6–9 h), in low-levels, an intense mezzo-scalar vortex over the
baroclinic zone (positive PV anomaly) that depleted it later in the
upper levels during the mature stage. The major development occurs
when the presence of the cyclonic vorticity in the upper levels
continues in spite of the diabatic mechanism tending to weaken it. The
surface latent heat fluxes and the Pontic marine features (SST
anomaly) seem to play a key role in the deepening of the storm and
a less important one during its mature phase. The aim of this paper is
to understand the synoptic and dynamic mechanisms relevant to this
development inside the Romanian geographical environment and to use it
further in similar cases. For meteorologists and operational weather
forecasters, it is therefore very important to recognise on the
satellite maps and different numerical model prognosis output this
phenomena features usually accompanied by high-impact and
meteorological risk.
Acknowledgements
The authors are indebted to Liviu Oana (West Univ. Timisoara, RoMetEx
organization) who contributed to the numerical simulations performed with the
WRF-ARW model in the frame of this study about the potential vorticity and
the surface heat flux. The Romanian National Meteorological Administration is
kindly acknowledged for providing radar and surface stations data used in
this paper. We are especially grateful to Ion Draghici for thought-provoking
discussions and professional advice. Gratitude is expressed to the
www.myocean.eu/ and the National Observatory of Athens for providing us
with SST and Sea Wave High analyses and also to all who supported and
encouraged us in this research.
References
Browning K. A.: The sting at the end of the tail: damaging winds associated
with extra-tropical cyclones, Q. J. Roy. Meteor. Soc., 130, 375–399, 2004.
Bruce, C. M. and Elmar, R. R.: Explosive cyclogenesis over the eastern United
States, Mon. Weather Rev., 116, 1568–1586, 1988.
Brunetti, M. and Capizzi, P.: Ciclogenesi sul Mediterraneo – Rivista di
Meteorologia Aeronautica, Anno 56, 1996.
Brunetti, M. and Moretti, A.: Le ciclogenesi del Mediterraneo, Un quarto di
secolo de osservazioni – Rivista di Meteorologia Aeronautica, Anno 65,
19–26, 2005
Conte M.: The meteorological “bomb” in the Mediterranean: a synoptic
climatology – Rivista di Meteorologia Aeronautica, Anno 46, 121–130, 1986.
Ion-Bordei, E.:
Rolul lantului alpinino-carpatic in evolutia ciclonilor mediteraneeni,
Printech Press, Bucharest, 82–91, 2009.
Gaza, R. S. and Bosard, L. F.:
Trough-merger characteristics over North America,
Weather Forecast.,
5, 341–331, 1990.
Gyakum, J. R. and Danielson, R. E.:
Analysis of meteorological precursors to ordinary and explosive cyclogenesis in the western North Pacific,
Mon. Weather Rev.,
128, 851–863, 2000.
Moore, R. W. and Montgomery, M. T.: Reexamining the dynamics of short-scale,
diabatic Rossby waves and their role in midlatitude moist cyclogenesis,
J. Atmos. Sci., 61, 754–768, 2004.
Neimann, P. J., Saphiro, M. A., and Fedor, L. S.:
The life cycle of an extratropical marine cyclone. Part II: Mesoscale structure and diagnostics,
Mon. Weather Rev.,
121, 2177–2199, 1993.
Nielsen, N. W. and Sass, B. H.:
A numerical, high-resolution study of the life cycle of the severe storm over Denmark on 3 December 1999,
Tellus A,
55, 338–351, 2003.
Parker, D. J. and Thorpe, A. J.: Conditional convective heating in a
baroclinic atmosphere: A model of convective frontogenesis, J. Atmos. Sci.,
52, 1699–1711, 1995.
Sanders, F. and Gyakum, J. R.:
Synoptic-dynamic climatology of the “bomb”,
Mon. Weather Rev.,
108, 1589–1606, 1980.
Santurette, P. and Georgiev, C. G.:
Weather Analysis and Forecasting – Applying Satellite Water Vapor Imagery and Potential Vorticity Analysis,
Elsevier Academic Press, Oxford, 2005.
Shapiro, M. A. and Keyser, D.:
Fronts, jet streams and the tropopause,
in:
Extratropical Cyclones. The Erik Palmen Memorial Volume,
edited by:
Newton, C. W. and Holopainen, E. O.,
Amer. Meteorol. Soc., Boston, MA, 167–191, 1990.
Strahl, J. L. S. and Smith, P. J.:
A diagnostic study of an explosively developing extratropical cyclone and an associated 500-hPa trough merger,
Mon. Weather Rev.,
129, 2310–2328, 2001.
The monthly and annual lowest station pressure (hPa) at Sulina (1961–2000).
The 25-year geographic distribution of the meteorological bomb
events (the number of cases) from 1980 to 2004 in the entire Mediterranean
Sea Basin (Brunetti and Moretti, 2005).
The “bomb” frequency events related to their intensity on the
Bergeron scale over the Mediterranean Basin from 1980 to 2004 (Brunetti and
Moretti, 2005).
Observed 36 h central pressure fall inside the cyclone (left) and
at Sulina weather station (right) from 09:00 UTC on 2 December to 21:00 UTC
on 3 December 2012.
Analysis of Δp in 24 h (black line, contours are labeled in 1 hPa) and NDR≥1 (green shade areas) at: (a) 03:00 UTC 3 December 2012, (b) 06:00 UTC 3 December 2012, (c) 09:00 UTC 3 December 2012, (d) 12:00 UTC 3 December 2012, (e) 15:00 UTC 3 December 2012, (f) 18:00 UTC 3 December 2012.
Analysis of 1000–500 thickness anomaly (shaded contours in dam) 3 day average ending on 3 December 2012.
Analysis of mean-sea-level pressure (contours are labeled in
1 hPa) at: (a) 12:00 UTC 2 December 2012 in the incipient
phase I, (b) 21:00 UTC 2 December 2012 in the phase II,
(c) 06:00 UTC 3 December 2012 in the explosive phase, (d)
18:00 UTC 3 December 2012 in the mature phase.
Analysis of 500 hPa geopotential height (solid line at 40 m intervals and of 300 hPa wind speed – shaded contours at 1 ms-1 values, greater than 25 ms-1 are shown) and surface locations (indicated by white circles) of the meteorological bomb at: (a) 00:00 UTC 2 December 2012, (b) 12:00 UTC 2 December 2012, (c) 00:00 UTC 3 December 2012, (d) 12:00 UTC 3 December 2012.
Analysis of the sea-level pressure (black solid lines at
3 hPa intervals) and of 500 hPa positive vorticity advection
(shaded contours in 10-8Ks-2) and of 500 hPa
temperature positive advection (brown contour lines in
10-4Ks-1) at: (a) 00:00 UTC 2 December 2012,
(b) 12:00 UTC 2 December 2012, (c) 00:00 UTC 3 December
2012, (d) 12:00 UTC 3 December 2012.
ECMWF analysis of 500 hPa geopotential height (lines at 40 m intervals) and of PV (magenta solid line in pressure height referred to 1 PVU) superimposed over Eumetsat IR images (ch.WV 6.3 micro) and surface locations (indicated by white circles) of the cyclone from 00:00 UTC 3 December in the explosive phase to 12:00 UTC 3 December 2012 in the mature phase.
Wind observations from the surface weather network stations during the storm.
Sequence of base velocity from the WSR-98 D Medgidia radar at 0.5 grades tilt at: (a) 00:24 UTC, (b) 02:48 UTC and (c) 05:23 UTC (left). The mask clouds, negative and positive wind flow speed and station plotting observations are identified. The radar distance coverage is about 500 km. Vertical cross-section from numerical model 00:00 UTC runs along the line AB at: (a) 00:00 UTC, (b) 03:00 UTC and (c) 06:00 UTC (right). The green-yellow shaped area is RH (70–120 %), the thin color lines represent the temperature (at 2.5 ∘C intervals), and the plotting barbs represent the wind direction and velocity; altitude is given in hPa. The relative location of the SJ is shown with a black circle.
The 60 h surface track (indicated by red squares) of the meteorological bomb from the incipient phase to the mature and dissipating phase (12:00 UTC 2 December–00:00 UTC 5 December 2012). Source: EUMETSAT RGB Channel at 09:00 UTC 3 December 2012.
Polar satellite images during the explosive phase of the cyclone. Source: NOAA-MODIS Channel 31/37 at 10:40 UTC 3 December 2012.
Observation analysis of the sea-level-pressure (red solid lines at 2 hPa intervals), of the surface moisture convergence (shaded colour contours, in gkg-1s-1), of the surface streamlines wind (green lines), and of the SST anomaly (shaded contours in Centigrades) at 03:00 UTC 3 December 2012. Model analysis of the potential vorticity anomaly (shaded colour contours) at 00:00 UTC 3 December 2012. “D” is similar to “L” from low pressure.
Observation analysis of mean-sea-level-pressure (red solid lines at 2 hPa intervals), of the surface moisture convergence (shaded colour contours, in gkg-1s-1), of the surface streamlines wind (green lines), and of the SST anomaly (shaded contours in ∘C) at 03:00 UTC 3 December 2012. Model analysis of the potential vorticity anomaly (shaded colour contours) at 03:00 UTC 3 December 2012. “D” is similar with “L” from low pressure.
Observation analysis of the mean-sea-level-pressure (red solid lines at 2 hPa intervals), of the surface moisture convergence (shaded colour contours, in gkg-1s-1), of the surface streamlines wind (green lines), and of the SST anomaly (shaded contours in ∘C) at 03:00 UTC 3 December 2012. Model analysis of the potential vorticity anomaly (shaded colour contours) at 06:00 UTC 3 December 2012.
Surface latent heat flux on 3 December at 07:00 UTC (left) and
10:00 UTC (right).