DYNAMO/AMIE S-PolKa Scientist Summaries - November 2011
 |
 |
 |
|
|
|
|
OTHER MONTHS:
|
| After one month of work in the
field, we pause to examine some statistics of the cloud
structure and vertical structure of the troposphere
compiled during the wet phase observed in mid to late
October, specifically 18 October through 28 October.
Statistics from the S-PolKa's S-band radar reveal
composite structures of precipitating areas that are
classified as convective or stratiform by an algorithm
that considers the horizontal structure of radar echoes.
Local peaks of reflectivity relative to surrounding echo
are classified as convective. The parameters of the
algorithm have been adjusted to minimize incorrect
classification of stratiform echo. Regions with strong
fall streaks are included in the convective category,
even if a weak bright band is present above. Statistics
are compiled separately for convective cells and
stratiform areas. Figure 1 and Figure 2 illustrate the vertical distribution of reflectivity detected by the S-band in two-dimensional probability density functions (PDFs) binned by height and reflectivity (hereafter referred to as CFADs). Areas classified as convective span a wide range of reflectivities in the boundary layer; although, the red contours in Figure 1 indicate that reflectivities between 20 dBZ and 30 dBZ are most commonly detected below the 0 deg C level. Above the 0 deg C level, the modal distribution extends toward lower reflectivities with increasing height until about 10 km. In the upper troposphere, the modal distribution of reflectivity is observed at 5 dBZ. A few convective cells have reached as high as 17 km. Echoes from regions classified as stratiform range from -5 dBZ to 30 dBZ in below the 0 deg C level. A robust bright band feature is detected at the 0 deg C level, denoted by a peak in reflectivity near that height. Above the 0 deg C level, reflectivities decrease to about 10 km and extend to the tropopause most often near 0 dBZ. The method that identifies convective precipitation areas is applied at low levels and does not take into account the slopes of echoes in sheared environments of the type frequently seen here, and the upper levels above convective areas identified in low-level echo patterns are included in the stratiform CFADs. Also, some of the regions classified as convective are shallow convective elements that remain below or penetrate partially through stratiform anvil at higher levels. Therefore, CFADs for convective and stratiform areas tend to be similar at upper levels above the 0 deg C level. Figure 3 shows CFADs for non-precipitating ice anvil clouds (those with no echo greater than -25 dBZ below 4 km) divided by their thickness. Thin anvils are those with depth less than 2 km, thick anvils have depths greater than 6 km, and the rest are medium anvils. Statistics are compiled using the cirrus mode from the DOE ARM Ka-band Cloud radar (KAZR) data during the period 12 October through 31 October. A large majority of non-precipitating anvils (81%) observed at Gan are thin anvils, and most of those occur near 12 km at -30 dBZ. Smaller peaks in the reflectivity distribution are observed near 8 km and -20 dBZ and 5 km and -30 dBZ. About 16% and 3% of anvils are medium and thick anvils, respectively, and their reflectivity distributions extend from about 12 km and -30 dBZ to near 0 dBZ at 6 km. Figure 4 illustrates the distribution of cloud top height for thin, medium, and thick anvils. The few thick anvils that do occur most often top out around 12 to 13 km. Medium and thin anvil tops are spread through a larger layer. While a peak in both is observed between 12 and 13 km, thin anvils most often top out just above the 0 deg C level. Some of this thin anvil results from shallow convection reaching a stable layer and depositing moisture. Other reflectivity returns are not connected to shallow convection and likely represent echoes enhanced by melting of small ice particles near that layer. Sounding data from the DYNAMO array (Male, Gan, Diego Garcia, and the R/V Revelle) are shown in Figure 5. Time series for zonal wind (u), meridional wind (v), and temperature (T) anomalies are shown at each site. Two separate vertically propagating modes are apparent in the lower stratosphere at each of the locations. A faster mode with a periodicity of about 4 days is seen in the time series for u at Male and Diego Garcia and v at Gan and Revelle. A slower mode with a periodicity of about 6 days in detected in v and T at Male and Diego Garcia and u and T at Gan and Revelle. Additional work is necessary to determine definitively what equatorial modes are observed. An analysis of anomalous geopotential may also reveal further information, including possible modes in the troposphere. Potential modes in the troposphere are interesting to explore since a 2 to 3 day cycle in convection has been observed during the recent active phase. Evidence of such frequency of convection during the moist phase can be seen in Figure 6, which shows daily rain accumulation derived from S-PolKa S-band. During the moist phase, a spike occurs about once every other day. A further look at hourly rain rates and convective echo tops (not shown) also strongly suggests that a diurnal cycle in convection occurs. After two consecutive days with lines of thunderstorms passing near Gan, we start November with numerous trade cumuli present in the area. Figure 7 is a photo taken around 06 UTC. Numerous trade cumuli are seen in the boundary layer, and some cirriform are also present. Occasionally, some clouds penetrated the 0 deg C level as illustrated in an RHI in Figure 8 from S-Polka S-band at 1454 UTC. Weather was quiet over most of the radar array as depicted in Figure 9. Some convection continued to persist near the Indian subcontinent, and more widespread deep convection and anvil in the eastern Indian Ocean, perhaps associated with a synoptic-scale feature, continued to propagate toward the east. Figure 10 shows the sounding from 1200 UTC at Gan. The dry layer near 700 mb that had persisted during the previous two days has moistened some, while northwesterly winds at that level have become lighter and more westerly. An 850 mb IMD WRF analysis at 00 UTC 1 November in Figure 11 continued to show westerlies extending across the Indian ocean to the south of a tropical cyclone near Oman and to the south of a weak low off the southwest coast of India. Looking foward, Figure 12 shows the new ECMWF monthly MJO forecast for the next 20 days. The ensemble mean suggests that only a weak MJO signal will propagate eastward through phases 4, 5, and 6, and considerable spread remains in the forecast beyond 10 days. However, the ensemble mean is suggestive that no active phase will occur during most of November; although, some of the members suggest that a weak MJO could initiate in the second half of November. |
| Figure 1. Two-dimensional PDF of S-band reflectivity distributed by height for columns classfied as convective during the period 18 October to 28 October 2011. The outermost dark blue contour represents 0.1%, and the contour interval is 0.2%. Reflectivity is binned by 5 dBZ, and height is binned by 0.5 km. The sum of all the values in this plot equals 1. |
 |
|
Figure 2.
Same as Figure 1 but for stratiform regions. |
 |
|
Figure 3.
CFADs for non-precipitating anvils as detected by KAZR
during the period 12 October through 31 October 2011.
Contour intervals and ranges are the same as in Figures
1 and 2. Here, for example, a red contour at -30 dBZ
and 12 km for thin anvils represents that 1.5% of all
anvils are thin anvils that have reflectivity of -30 dBZ
at 12 km. |
 |
|
Figure 4.
Probability density function of cloud top height for
anvils detected by KAZR. |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
Figure 5.
Time series of anomalies of zonal wind, meridional wind,
and temperature at Male, Gan, Diego Garcia, and
the Revelle. Revelle moved off its
station at the equator in late October. |
 |
|
Figure 6.
Total daily rain accumulation within 150 km of S-PolKa
S-band during the period 5 October to 31 October. |
 |
|
Figure 7.
Photo taken looking north from S-PolKa around 06 UTC 1
November 2011. Courtesy: Bob Houze |
 |
|
Figure 8.
An RHI taken at an azimuth angle of 52 degrees at 1454
UTC from S-PolKa S-band. Reflectivity is shown
here. |
 |
|
Figure 9.
METEOSAT infrared satellite imagery from 1200 UTC 1
November 2011. |
 |
|
Figure 10.
Sounding from Gan for 1200 UTC 1 November 2011. |
 |
|
Figure 11.
850 mb analysis for IMD WRF-NMM 27 km run at 00 UTC 1
November 2011. |
 |
|
Figure 12.
ECMWF MJO 20-day forecast generated on 31 October
2011. |
|
The latest Cooperative Institute for Climate and
Satellites (CICS) projection of Outgoing Longwave
Radiation Anomalies in Figure 1
strongly indicates that we are going into a dry period in
the DYNAMO array. The DOE Gan sounding at 0000 UTC in Figure 2, however, shows that the
entire troposphere remains relatively moist (recall how
dry it was in the last suppressed phase ~10 October).
Infrared imagery in Figure 3 shows
relatively small amounts of high cloud tops in the region.
The few high cloud tops in the region were in the form of
elongated streaks oriented NE-SW, in the direction of the
upper level winds seen in Figure
2. Figure 4 shows that only
a few convective-scale echoes were present on the SMART-R
C-band radar (similar for S-PolKa). From the ground, the
clouds were seen to be in the form of small and towering
cumulus, all of which were growing vigorously, appearing
quite healthy (Figure 5, left and
middle). In the distant SE the tops of cumulonimbus were
evident (Figure
5, middle). One of the cells producing precipitation
is seen in the right panel of Figure
5. We have noticed a local tendency for such storms
to form to the east. Considerable shipping occurs coming
into the port just east of the S-Pol site and plumes of
smoke from burning of garbage on atoll are all usually
upstream of where this storm was occurring. This type of
storm we have dubbed "the port storm" and is probably not
of any large scale significance but rather more of a local
curiosity, possibly related to pollution. The cumulus and
small cumulonimbus in the area were seen on the S-PolKa
radar to be triggered at the boundaries of echo-free cold
pools, as we have noted in several previous summaries. Figure 6 shows typical radar echo
structures seen during the day. Note from the bottom right
panel that the ZDR data during the day were not showing a
positive signal at the cold pool boundaries. We have
speculated that the ZDR positive signal in gust fronts
might occasionally be birds (see previous summaries). That
positive ZDR in the boundary layer might be birds is
further suggested by Figure 7, which
shows patches of positive ZDR (red) emanating from the
island at moving out to sea between 5 and 6 o'clock in the
morning local time on this day. Radial velocity patterns
looked noisy in these zones of red ZDR. Late in the day an anvil from one of the elongated NE-SW oriented streaks seen in the infrared data was over Gan. Figure 8 shows three views of the anvil: an infrared image, particle type signals from the S-PolKa polarimetric variables, and the reflectivity seen by the DOE KAZR. The anvil was relatively weak, with one strong fallstreak, likely from dying embedded convection. |
| Figure 1.
Latest Cooperative Institute for Climate and
Satellites (CICS) projection of Ougoing Longwave Radiation
Anomalies. The vertical dashed lines are at the longitude
of the DYNAMO array. |
| Figure 2. DOE Gan
sounding for 0000 UTC 2 November 2011. |
 |
| Figure 3. Selected
METEOSAT infrared images for 29-30 October 2011. |
 |
 |
| Figure 4. SMART-R
C-band reflectivity observations at Gan superimposed on
satellite imagery. |
 |
 |
 |
| Figure 5. Photos take
at the S-PolKa site facing NE at 0616 UTC (left) and ESE
(middle) and E (right) at 0717 UTC 2 November 2011. |
 |
 |
|
 |
| Figure 6. SMART-R
C-band reflectivity observations at Gan superimposed on
satellite imagery. |
 |
 |
 |
| Figure 7. S-PokKa ZDR
fields for 0016-0046 UTC (0516-0546 local) 02 November
2011. |
 |
||
|
||
|
||
| Figure 8. METEOSAT infrared image, S-PolKa particle type field, and DOE KAZR data in an anvil over Gan at about 1800 UTC 2 November 2011 |
| The moist disturbed region of the
last few weeks is moving off to the east, and today we
experienced some last deep convection on the back edge of
this region. This convective event was rather highly
electrified. The 850 hPa map in Figure
1 shows a cyclonic shear zone extending from about
5S and 50E to Gan and farther to the east. Westerly flow
just north of this feature was directed from Africa toward
Gan. The DOE Gan sounding showed a generally moist
troposphere (Figure 2). The METEOSAT
infrared imagery in Figure 3 shows a
band of convective elements with elongated anvils
distributed roughly parallel to the shear line seen in the
850 hPa winds pattern. Figure 4
shows that lightning was strongly associated with the
cells located along the shear line. The overlaid fields in
Figure 5 show that the S-PolKa
observed one of these convective elements. The echo fields
seen in Figure 6 show that the
leading edge of the convective element was highly
convective, reaching 16 km and lofting graupel to 8 km and
other large ice particles to even higher levels. This
microphysical structure is consistent with the occurrence
of lightning in convective features similar to this one
observed by S-PolKa. The PPI patterns in the upper tier of
Figure
6 show further that the leading convective elements
were trailed by stratiform precipitation falling from the
plumes shearing off to the west in the upper level
easterlies. These convective elements lacked the ability
to grow upscale into large mesoscale systems. Later in the
day, one of the elongated convective features was observed
from below from the S-PolKa site. Figure
7 shows views of this system with the convective
elements seen in the distant east. Altostratus was under
the cirrostratus anvil (left panel) and billow clouds
possibly formed by small-scale gravity waves were
emanating from the anvil at various locations along its
length (right panel). Figure 8
shows photos from a similar perspective but when the anvil
was less active. |
| Figure 1.
Indian Meteorological Department 0-hour analysis of 850
hPa winds for 0000 UTC 3 Novemeber 2011. |
| Figure 2. DOE Gan
sounding for 0300 UTC 3 November 2011. |
 |
|
| Figure 3. Selected
METEOSAT infrared images for 3 November 2011. |
| Figure 4. World Wide
Lightning Location Network flashes for the previous 30 min
superimposedo the METEOSAT infrared image for 0230 UTC 3
November 2011. |
|
 |
| Figure 5. SMART-R
C-band reflectivity observations at Gan superimposed on
satellite imagery, 0209 UTC 3 November 2011. |
|
|
|
 |
 |
 |
| Figure 6. SMART-R
C-band reflectivity observations at Gan superimposed on
satellite imagery. |
|
| Figure 7. Photos
looking E (left) and ESE from the S-PolKa site at 0741 UTC
3 November 2011. |
|
| Figure 8. Left to
right: Photos taken at the S-PolKa site at 1559 UTC 3
November 2011..Left: Anvil source in the distant east and
extending overhead. Middle: Side edge view of the anvil
directly overhead. Right: Anvil base extending to the
west. |
| Westerly flow at and north of the
equator at the longitude of the DYNAMO array prevailed
throughout the day at the 850-500 hPa levels (Figures 1, 2 and 3). At 700 hPa, the westerlies originate
near Africa. At all levels, they lie between cyclonic
circulations both north and south of the equator in the
western Indian Ocean. The DOE Gan soundings show the
humidity decreasing in mid-to-high levels (Figure 4). The sounding at 0000 UTC
suggests a correlation of the strength of the westerlies
at 650 and 500 hPa with the dryness at those levels. The
METEOSAT infrared image in Figure 5
shows the main cloudiness over the Maritime Continent with
suppressed conditions over the southern DYNAMO arrray.
Some high cloudiness is located south of the equator near
Diego Garcia, at the southern extremity of the DYNAMO
array. Despite the dry and suppressed conditions, some
precipitation echo cells appeared on the S-PolKa S-band
and SMART-R C-band radars during the heat of the day.
Precipitation echoes were absent during the night. The
general echo sequence is seen in Figure 6. At 0400 UTC (top
panel) no precipitation echoes were present. During the
day (subsequent panels) the echoes formed east of the
radars, grew, and moved eastward. On the S-PolKa radar,
the echoes formed in locations
of non-precitating cloud lines, also evident in the
weaker reflectivites seen in low reflectivity echoes.
Examples of the echoes forming out of these cloud lines
can been seen 105 km NNE and 60 km ESE of the radar in the
0807 UTC panel of Figure
6. A more detailed sequence is shown in Figure 7. Numerous cloud line echoes
produced by Bragg scattering can be seen in the PPIs east
of the radar during the half hour time period shown. At
the end of the period, a precipitating cell was growing
out of a line where the yellow line crosses ~80 km range.
The RHIs in the bottom tier of the figure, taken along the
yellow line, show non precipitating clouds reaching 1-3
km, with those approaching 3 km producing stronger echoes
as they start precipitating. Figure 8
shows photos of the ubiquitous smaller nonprecipitating
clouds occurring in the vicinity of S-PolKa throughout the
period of about 0500-1200 UTC. Occasionally, one of the
nonprecipitating clouds grew into a cumulus congestus or
isolated cumulonimbus. Figure 9
shows several examples of these taller clouds seen between
about 0700 and 1200 UTC. The cumulonimbi in the upper
right of Figure 9 produced the
long extended anvils seen in the lower left panel. |
| Figure 1.
Indian Meteorological Department 0-hour analysis of 500
hPa winds for 0000 UTC 4 November 2011. |
| Figure 2.
Indian Meteorological Department 0-hour analysis of 700
hPa winds for 0000 UTC 4 November 2011. |
| Figure 3.
Indian Meteorological Department 0-hour analysis of 850
hPa winds for 0000 UTC 4 November 2011. |
|
|
| Figure 4. DOE Gan
soundings for 4 November 2011. |
 |
| Figure 5. METEOSAT
infrared image for 0730 UTC 4 November 2011. |
|
 |
 |
 |
| Figure 6. SMART-R
C-band (left) and S-PolKa S-band (right) reflectivity
patterns superimposed on METEOSAT water vapor channel
image for 0400-1000 UTC 4 November 2011. |
|
|
|
 |
 |
 |
| Figure 7. S-PolKa
S-band reflectivity patterns for 0431-0510 UTC 4 November
2011. |
|
|
|
| Figure 8. Photos taken at the S-PolKa site on 4 November 2011. Upper row: 0452 UTC (left); 0544 UTC (right). Middle row: 0653 UTC (left); 0852 UTC (right). Bottom row: 1143 UTC. |
|
|
| Figure 9. Photos
taken at the S-PolKa site on 4 November 2011. Clockwise
from upper left: 0652, 0724, 0857, 1145 UTC. |
| The drying and suppressed
conditions noted yesterday were even more pronounced
today. The 700 hPa map in Figure 1
shows westerly wind and low relative humidity along the
equator from Africa to the DYNAMO array. The DOE Gan
soundings in Figure 2 show the
relative humidity above the 850 hPa level decreasing
between 0000 and 0600 UTC and lowest between 0600 and 0900
UTC (1100-1400 local time). The 700 hPa level showed a
minimum with westerlies at that level, consistent with the
synoptic pattern seen in Figure
1. The METEOSAT infrared image in Figure 3 shows no clouds in the
region surrounding Gan and also very little cloudiness in
the visible image. Photographs taken throughout the day (Figure 4) showed mostly very small
cumulus clouds, generally struggling to survive. Though
the boundary layer remains moist, as always, the clouds
were having trouble penetrating the dry layer just above
the boundary layer. The cloud in the third row left of
Figure 4 is a typical example of a
cumulus rising into the dry layer and having its top
evaporate. The S-PolKa S-band
PPIs in Figure 5 show mostly
non-precipitating clouds highlighted as a result of Bragg
scattering. The RHIs in the lower row of Figure 5 show cross sections
indicating that the low clouds
were almost all confined to the layer below 3 km (i.e. the
700 hPa level where the intense drying was seen in the
soundings in Figure 2). By 1100
UTC (1600 local time) a few isolated
cells were penetrating to about 5 km. Figure
6 shows one cell reaching 5.5 km and even triggering
some heavy rain signatures in the polarimetric particle
identification algorithm (bottom middle panel). |
| Figure 1.
Indian Meteorological Department 0-hour analysis of 700
hPa winds for 0000 UTC 5 November 2011. |
|
|
|
|
| Figure 2. DOE Gan
soundings for 5 November 2011. |
 |
 |
| Figure 3. METEOSAT
infrared image for 0730 UTC 4 November 2011. |
 |
|
|
|
| Figure 4. Photos
taken at the S-PolKa site on 5 November 2011. Top row:
0521, 0648. Second row: 0846, 1010. Third row: 1132, 1208.
Bottom row: 1246 UTC. |
|
|
|
 |
 |
 |
| Figure 5. S-PolKa
S-band reflectivity patterns seen between 0245 and 1130
UTC 5 November 2011. |
|
||
 |
 |
 |
| Figure 6. S-PolKa
S-band data at 1200 UTC 4 November 2011. |
| Dry air continues to be advected
into the northern part of the DYNAMO region at 700 hPa (Figure 1). The DOE Gan sounding
continues to show an especially dry layer and westerly
winds in the 700 hPa level (Figure 2).
It is also very dry in upper levels (from 350 hPa
upwards). The METEOSAT inrared image over the whole region
shows the southern DYNAMO array in a zone with no major
cloud systems (top panel of Figure 3).
Zoomed in (middle panel) the image shows a few
cumulonimbus tops reaching the 208 K level. The visible
image in the bottom panel shows isolated cumulus clouds
scattered through out the region east of Gan. The photos
in Figure 4 show the clouds near
the S-PolKa site at various times throughout the day. The
upper left photo, taken before mid-morning, shows only
very small cumuli in the foreground. The anvils of the
taller cumulonimbi to the south are visible in the
distance. The lower left image, taken over 3 h later again
shows an anvil in the distant south and the remains of an
anvil closer to the radar site. The right-hand photos show
that while most of the cumuli were small, a few grew into
towering cumulus congestus. These clouds looked somewhat
more vigorous than yesterday's clouds, but the difference
was small at best. In the distance in the right-hand
images, towers of congestus can be seen trying to rise a
bit higher but withering away as the tops are sheared
off. An interesting feature that we have been
noticing on radar is what appear to be Bragg scattering
reflections from layers of strong gradients of humidity.
Such features have been described in a recent Radar
Conference paper by Jennifer Davison and others and are
part of her Ph. D. studies at the University of Illinois.
As found by her, we are seeing layers of Bragg scattering
associated with the lower-level moist layer. Figure 5 shows PPIs at 9 deg
elevation. The center of the beam at the 25 deg range ring
is at about 5 km altitude. The left hand panel of Figure 5 shows that from 4.5 km
downward (within ~20 km of
the radar in these 9 deg elevation PPIs), contiguous
layers of low reflectivity were producing a series of echo
rings (at different altitudes). No layer clouds could
account for these rings since none were present at this
time. Above these levels, the water vapor content was too
low to produce such reflections. The middle panel shows
that the the rhoHV signal (correlation between vertically
and horizontally polarized returns) is near zero, and the
ZDR signal in the right-hand panel is noisy. These signals
indicate that the turbulent elements producing the
scattering are inhomogeneous. Another example is in the
lower left panel of Figure 6. The
somewhat stronger reflections seen at lower levels (~-5
to +3 dBZ) at ranges ~3-7 km, with bases near 0.5 km and
tops at about 1.2 km, are from the non-precipitating
cumulus clouds in the vicinity of the radar, as seen in
the photos in Figure 4. The lower
middle panel of Figure
6 shows a view illustrating the way in which these
cumulus cloud echoes are also associated with Bragg
scattering, but differently than in the moisture layers.
The rhoHV signal in the clouds forms a red "mantle" around
the the cloud. The red corresponds to correlations near
1.0, indicating reflection from turbulent elements in the
moisture gradient zone at the cloud edges. These elements
have a rhoHV of 1.0, indicating a rather homogeneous
collection of scatterers. This type of mantle echo is well
known to be produced by Bragg scattering. Close
examination of the ZDR signal shown in the lower right
panel of Figure
6 shows near zero values in these cumulus clouds,
also consistent with the scatterers having no particular
spatial orientation. In the early afternoon (~0830 UTC,
1330 local), some isolated precipitating echoes had
formed. The lower left panel of Figure
7 shows that some of these precipitating cloud
elements had tops that reached 8 km. The lower middle
panel again shows the moisture layer signals produced by
Bragg scattering as well as a cumulus cloud near the radar
that was begining to produce a Rayleigh scattering signal
from drops (~9 dBZ). The particle type algorithm
classifies these drops as "drizzle," i.e. the smallest
precipitating drop size category. The more intense cell
near 120 km range contained some heavy precipitation in
its core and was glaciating near its top. A few isolated
cells with similar characteristics continued to form until
later in the afternoon (Figure 8),
but these clouds in view of S-PolKa could not survive
rising through the dry layer beyond about 8 km height. |
| Figure 1.
Indian Meteorological Department 0-hour analysis of 700
hPa winds for 0000 UTC 6 November 2011. |
|
|
| Figure 2. DOE Gan
soundings for 6 November 2011. |
 |
 |
|
| Figure 3. METEOSAT
infrared images for 0900 UTC 6 November 2011. |
 |
|
| Figure 4. Photos
taken at the S-PolKa site on 6 November 2011. Top row:
0511 UTC (facing south) and 0708 UTC (facing east). Bottom
row: 0828 UTC (facing south) and 0941 UTC (facing east). |
 |
 |
 |
| Figure 5. S-PolKa
S-band PPIs data taken at elevation 9 deg at 0305 UTC on 6
November 2011. Left to right: reflectivity, rhoHV, ZDR.
Note range ring is at 25 km. |
|
||
 |
 |
 |
| Figure 6. S-PolKa
S-band data taken between 0531 and 0539 UTC on 6 November
2011. Upper: PPI of reflectivity. Lower: RHIs of
reflectivity, polarimetric correlation coefficient
(rhoHV), and differential reflectivity (ZDR). |
|
||
 |
 |
 |
| Figure 7. S-PolKa
S-band data taken at about 0830 UTC on 6 November 2011.
Upper: PPI of reflectivity. Lower: RHIs of reflectivity
(unzoomed), reflectivity (zoomed), and polarimetrically
determined particle type. |
|
||
 |
 |
|
| Figure 8.
S-PolKa S-band data taken at about 1020 UTC on 6 November
2011. Upper: PPI of reflectivity. Lower: RHIs of
reflectivity (unzoomed), reflectivity (zoomed), and
polarimetrically determined particle type. |
| This day was a continuation of the
convectively suppressed conditions we've seen the last few
days. The easterlies at 200 hPa are strong north of the
equator, but drop off in intensity south of the equator
over the DYNAMO four-corners array (Figure
1). Weak northwesterlies at 700 hPa continue to
advect dry air toward Gan (Figure 2).
However, the DOE Gan soundings show generally weak winds
from 700 to 400 hPa (Figure 3). The
soundings also show dramatically increased drying
conditions aloft, reminiscent of the last suppressed phase
on about 10 October (see previous summaries). The METEOSAT
infrared image over the Indian Ocean shows some minor
cumulonimbus activity in the four-corners array, south of
Gan (Figure 4, first two panels).
The third panel of Figure
4 contains a visible image, which shows isolated
cumulus activity east of Gan. Such clouds dominated the
days cloud population in the vicinity of S-PolKa. Before
referring further to the days cloud population, it's
interesting to note again the behavior of the moist
boundary layer under these convectively suppressed
conditions (see also summaries for various days in
October). Figure 5 shows low-elevation
PPIs for two times during the night (left and middle
panels) and one time after sunrise (right hand panels). At
the first time a precipitation echo was located about 40
km due east of the radar (left panels). Much of the area
within about 60 km of the radar was covered by "clutter"
echoes. A little over three hours later (middle panels),
an empty "hole" occurred in the clutter where the rain
cell had been located. We usually interpret these holes as
"cold pools" left behind by rainshowers, and numerous
examples could be seen over night and into the daytime.
Another fascinating behavior is the change in the ZDR
signature from nighttime to daytime (lower tier of
panels). At night the clutter patterns have a predominance
of red colors, which correspond to positive ZDR values,
while in the daytime the ZDR is more nearly averages to
zero (right lower panel). The reason that the clutter
disappears in the cold pools is likely related to the
vertical gradient of humidity being different in the cold
pools compared to the undisturbed boundary layter. The
reason for the more positive ZDR in the boundary layer at
night awaits explanation but may also be related to the
vertical gradient of humidity changing between night and
day. Figure 6 shows clouds seen
from the S-PolKa site at various times of day. In the
lower right panel of Figure
6, the tops of higher clouds in the distant south
coud be seen; they are the cumulonimbi seen in the middle
of the four-corners array in Figure
4. Clouds more local to S-PolKa were genrallly
sparse, but occasionally growing into cumulus congestus.
When they tried to go to higher levels (as in 2nd row
right or 3rd row left), the tops just evaporated.
Occasionally, however, the congestus clouds precipitated
robustly. One such cloud formed and rained out close to
S-PolKa (2nd row photo in Figure
6). The radar obtained in this congestus shower are
shown in Figure 7. Despite the
cloud's small size, it produced a 50 dBZ echo (top and
bottom left left) and heavy rain was indicated by the
particle type algorithm (middle bottom). The cloud showed
no signature of glaciation. This small shower illustrates
how much water (and latent heat) can be squeezed out of
the moist layer with no contribution from ice phase
microphysics. The radial velocity data indicated that a
downdraft produced a divergent signature near the surface
(upper right). Some evidence of divergence is seen near
the top of the echo in the radial velocity RHI (lower
panel). |
| Figure 1.
Indian Meteorological Department 0-hour analysis of 200
hPa winds for 0000 UTC 7 November 2011. |
| Figure 2.
Indian Meteorological Department 0-hour analysis of 700
hPa winds for 0000 UTC 7 November 2011. |
|
|
| Figure 3. DOE Gan
soundings for 7 November 2011. |
 |
 |
|
| Figure 4. METEOSAT
infrared images for 0800 UTC 7 November 2011. |
 |
 |
 |
|
|
|
| Figure 5. S-PolKa
S-band PPIs data taken at elevation 9 deg on 7 November
2011. Upper panels show the reflectivity field. Lower
panels show ZDR. The left and middle columns are for
nighttime conditions: 1846 and 2201 UTC, corresponding to
2346 and 0301 local time. The right column is for 0216 UTC
or 0716 local time (an hour and a quarter after sunrise). |
 |
|
|
| Figure 6. Photos
taken at the S-PolKa site on 7 November 2011. Top row:
0412 UTC (facing east) and 0558 UTC (facing east). Second
row: 0634 UTC (facing east) and 0820 UTC (facing north).
Bottom row: 0927 UTC (facing northeast) and 1131 UTC
(facing southeast). |
|
||
|
|
|
| Figure 7. S-PolKa
S-band PPIs data taken at about 0630 UTC on 7 November
2011. Upper row, left to right: PPIs of reflectivity and
radial velocity at 0.5 deg elevation. Note range ring is
at 25 km. Lower row, left to right: RHIs of
reflectivity, particle type, and radial velocity along
the yellow line in the upper panels. |
| Today was another quiet day in
Gan. Figure 1 shows an example of
one of the deepest cells to develop within the radar
domain. Most cells topped out at most between 3 and 3.5
km; although one area of deep convection extended at least
to 15 km and was observed in the far southeastern portion
of the domain between 0900 UTC and 1000 UTC (not shown). A
view of reflectivity in early afternoon is seen in Figure 2, and the yellow line
represents the cross section of cumulus clouds depicted in
Figure 3. Such shallow boundary layer
clouds were typical for the suppressed day. Figure 4 is the 1200 UTC atmospheric sounding taken at Gan, and it shows that much of the troposphere was rather dry. A time series of vapor density deviation from the time-mean since 1 October is shown in Figure 5. During the past few days, significant drying has occurred throughout the troposphere above 3 km; however, the lower troposphere below ~800 hPa remains moist. Westerly flow from Africa and around the southwest side of a cyclone in the Arabian Sea have transported dry air into the area, and Figure 6 illustrates the result by showing total precipitable water over the Indian Ocean derived by the University of Wisconsin-CIMSS from AMSR-E and SSMI. Figure 7 illustrates the growth of a convective ring that formed near and to the west of Diego Garcia on 7-8 November. Late on 7 November, deep convection developed west of Diego Garcia. On 8 November, the area of high, cold cloud tops expanded over a large area. By mid-afternoon and evening, the areas of deepest convection had extended laterally outward in all directions from the area of initial deepening, and only low clouds remained near the center of the apparent convective ring. Various cumulus cloud lines developed throughout the day. Figure 8 is a visible satellite image taken at 1030 UTC which demonstrates the orientation of such lines. The clouds in the western portion of the radar domain were oriented meridionally, and the clouds in the eastern portion of the domain were more zonally oriented. However, all of the clouds propagated toward the south. The zonally oriented clouds may have been lined up along flow around a developing area seen several hundred kilometers northeast of the radar site, while the north-south oriented clouds formed bands that were advected into the disturbance west of Diego Garcia. A radar image at 1030 UTC is seen in Figure 9. Most of the deeper cells that did form, such as the previous echo at 15+ km, developed south of S-PolKa along the zonally oriented lines. |
| Figure 1. A
reflectivity RHI of a cumulus cloud at an azimuth angle of
141.9 at 0930 UTC 8 November. |
 |
| Figure 2. Reflectivity
observed by S-PolKa S-band radar at a 1.5 deg elevation
angle at 0646 UTC 8 November. |
 |
| Figure 3. Cross
section of reflectivity for a line of cumulus clouds shown
in Figure 2 at 0650 UTC 8 November. |
 |
| Figure 4. DOE sounding
taken at Gan at 1200 UTC 8 November. |
 |
| Figure 5. Time series
of deviation of vapor density from the time-mean at Gan.
Time-means are calculated by 100 meter bins in height
using data since October 1. |
 |
| Figure 6. Analysis by
the University of Wisconsin of total precipitable water
over the Indian Ocean using AMSR-E and SSMI. |
 |
 |
 |
 |
| Figure 7. METEOSAT
visible satellite imagery taken at 1900 UTC 7 November
(upper left), 0200 UTC 8 November (upper right), 0930 UTC
8 November (lower left), and 1630 UTC 8 November (lower
right). |
 |
| Figure 8. METEOSAT
visible satellite imagery taken at 1030 UTC 8 November. |
 |
| Figure 9. Reflectivity
from S-PolKa S-band at 1030 UTC 8 November. |
| The S-PolKa has been on the northern fringe of strong convective activity south of the equator for the past two days. At 200 hPa, the winds are southeasterly in the northeastern quadrant of an anticyclonic circulation, with stronger easterlies lying just to the north of the equator (Figure 1). At 500 hPa the winds around Gan are north-northeasterly (Figure 2). At 850 hPa, the winds are northwesterly (Figure 3). All three of these wind directions are representive of layers seen in the DOE Gan soundings at 0600 and 0900 UTC (Figure 4). The major convective event of this day close to the DYNAMO array was another giant ring of convection, similar to what we have described on previous days (see 22 October, 24 October, 8 November summaries). A sequence of METEOSAT infrared images at 3 h intervals in Figure 5. The sequence begins on 8 November at 0800 UTC and shows again the ring event on that day. It started with a mesoscale convective outbreak at about 7S, 65E (top left panel). Following the sequence we see the ring of convective systems expanding in a ring and leaving a large relatively open center at 1200 UTC 8 November (2nd row left). At about 1800 UTC a large convective system over Diego Garcia formed the center of another expanding ring, culminating in the circular pattern of convection seen at 0600 and 0900 UTC in the last two panels of Figure 5. The convection in the ring was highly electrified (Figure 6). By 0900 UTC the ring was weakening (last panel of Figure 5). The pronounced northwestward streaming anvils on the northern side of the ring were consistent with the maximum of easterlies seen along 2.5 S between 60 amd 75 E in Figure 1. The intense convection of the ring was beyond the range of S-PolKa. However, all day cirrostratus clouds were overhead, streaming away from the ring (left hand photo in Figure 7). These layer clouds were thick and seen on the DOE Gan KAZR to have bases ranging from upwards of 7 km (Figure 8). The KAZR Doppler velocity was mostly downward at about 1 m/s throughout the layer cloud, although a couple cells of upward velocity were detected in the upper part of the cloud layer (lower panel of Figure 8). Cumulus congestus clouds were occurring below the upper anvil cloud layer (right hand photo in Figure 7). The KAZR detected updraft in one such cloud at about 2100 UTC on 8 November. Isolated convective cells occurred in the eastern half of the S-PolKa area during the day (Figures 9 and 10). They were generally associated with cold pool triggering, and were somewhat deeper than the precipitating cloud elements we have seen in the last few days. These cumulonimbi were getting about the 0 deg C and producing ice, as can be seen in the lower right panels of Figures 9 and 10. |
| Figure 1.
Indian Meteorological Department 0-hour analysis of 200
hPa winds for 0000 UTC 9 November 2011. |
| Figure 2.
Indian Meteorological Department 0-hour analysis of 500
hPa winds for 0000 UTC 9 November 2011. |
| Figure 3.
Indian Meteorological Department 0-hour analysis of 850
hPa winds for 0000 UTC 9 November 2011. |
|
|
| Figure 4. DOE Gan
soundings for 9 November 2011. |
 |
|
|
|
|
|
|
|
|
|
|
|
| Figure 5. METEOSAT
infrared images for 8-9 November 2011. |
 |
| Figure 6. World Wide
Lightning Location Network data for 30 min preceding 30
min superimposed on METEOSAT infrared image for 0900 UTC 7
November 2011. |
 |
| Figure 7. Photos
taken by Kaustav Chakravarty at the S-PolKa site at about
1100 UTC 9 November 2011, facing west (left) and south
(right). |
 |
 |
| Figure 8. Data from
the DOE Gan vertically pointing Ka-band KAZR radar
data for 8-9 November 2011. |
|
||
|
|
|
| Figure 9. S-PolKa
S-band 0.5 deg PPI at 0709-0716 UTC 9 November 2011
(upper) and vertical RHIs along the yellow line of
reflectivity (lower left) and hydrometeor identification
(lower right). |
|
||
|
|
|
| Figure 10. S-PolKa
S-band 0.5 deg PPI at 0912-0916 UTC 9 November 2011
(upper) and vertical RHIs along the yellow line of
reflectivity (lower left) and hydrometeor identification
(lower right). |
| Major convective activity occurred
today northwest of the southern DYNAMO four-corners array.
Fortuitously the Revelle passed through that region of
convection during its transit back to the array and
obtained both sounding and radar data in the region of
active convection. S-PolKa's area of coverage touched the
southwestern end of the region of convection, and today
the convection see by S-PolKa was somewhat deeper than in
preceding days, during which we've seen deeper convection
each day, although none is yet growing upscale into
mesoscale systems. The 200 hPa flow remains easterly along
and north of the equator between large-scale anticyclonic
patterns north and south of the equator (Figure
1). The 700 hPa winds are confluent dry westerlies
over the four-corners southern array (Figure
2). At 925 hPa, the flow is weak over the array,
which lies in a col region (Figure 3).
The DOE Gan soundings in Figure 4
are consistent with this pattern, and further show weak
southeasterly wind near the surface. The major convective
event of the day was northeast of the four corners array.
Its history is shown in Figure 4.
Figure 5 shows that the convection in
this region was most intense about 0100-0200 UTC, and the
convective area was elongated with a SW-NE orientation,
with the convective region expanding somewhat like the
rings discussed previously in these summaries. The S-Polka
site lay between this region and the region of active
convection centered at about 7S, 63E. Figure 6 shows that the active
region to the northeast of the array was not very active
electrically, with just a few detected flashes. The Revelle was on a
transit from Pukhet to the NE position of the array and
passed directly through this zone of active convection.
Three Revelle
soundings taken during the transit through the convective
region (Figure 7) show the
moist conditions expected in this region. The 0000 UTC
sounding was truncated ner the 0 deg C level, which often
happens in stratiform precipitation regions due to ice on
the balloon. Note the strong southwesterly winds at low
levels in the soundings taken in the disturbed region,
consistent with the 925 hPa wind analysis in Figure
3. Figure 8
shows an example of the mesoscale radar echoes within
the region seen on the Revelle radar during the ship's passage
through the convective zone. A combination of convective
lines with various orientations and large stratiform
rain areas were seen all along the transit. Back at
S-PolKa, the situation was much less disturbed but
convective cells were stronger than in
the last few days. The photos in Figure
9 show several views of the cumulus clouds growing
in the vicinity of S-PolKa. A large cirrostratus overhang
advected into the region from more convectively active
regions by upper level winds were present all day. The DOE
Gan KAZR data show the bases of this layer rising from 6
to 10 km through the day (Figure 10).
The cumulus and cumulonimbus cells that gained vertical
development in the local area extended upward through and
interacted with the upper cloud shield. The photo in the
upper left of Figure
9 shows an example seen visually. The KAZR data in Figure 10 show and example at about
0900 UTC. The S-PolKa S-band data in Figure
11 show an example seen on radar; a narrow
convective cell is seen extending up through the
cirrosratus layer. In Figure 12,
the convection has built a thicker stratiform ice cloud
with the beginning of a melting layer apparing in the
particle type cross section. Later in the day, to the
north, convective cells were seen reaching about 14 km
with large heavy rain and graupel and with larger ice
particles being lofted to high levels
(Figure 13). This structure
is similar to what we have seen repeatedly in the deeper
convective cells. |
| Figure 1.
Indian Meteorological Department 0-hour analysis of 200
hPa winds for 0000 UTC 10 November 2011. |
| Figure 2.
Indian Meteorological Department 0-hour analysis of 700
hPa winds for 0000 UTC 10 November 2011. |
| Figure 3.
Indian Meteorological Department 0-hour analysis of 925
hPa winds for 0000 UTC 10 November 2011. |
|
|
| Figure 4. DOE Gan
soundings for 9 November 2011. |
 |
|
|
|
|
|
| Figure 5. METEOSAT infrared images for 8-9 November 2011. |
 |
| Figure 6. World Wide
Lightning Location Network data for 30 min preceding 30
min superimposed on METEOSAT infrared image for 0230 UTC
10 November 2011. |
|
|
|
| Figure 7. Revelle soundings for
9-10 November 2011. |
| Figure 8. Revelle radar PPI for
0159 UTC 9 November 2011. |
 |
|
|
| Figure 9. Photos
taken by Kaustav Chakravarty at the S-PolKa site at: Top
row, 0430 UTC 9 November 2011, facing west (left) and
southeast (right). Middle row: facing east, 0715 UTC
(left) and 0930 UTC (right). Bottom row: facing southeast
at 1050 UTC. |
 |
| Figure 10. Data from
the DOE Gan vertically pointing Ka-band KAZR radar
data for 9-10 November 2011. |
|
||
|
|
|
| Figure 11. S-PolKa
S-band 0.5 deg PPI at 0845-0900 UTC 10 November 2011
(upper) and vertical RHIs along the yellow line of
reflectivity (lower left), hydrometeor identification
(lower middle), and radial velocity (lower right). |
|
||
|
|
|
| Figure 12. S-PolKa
S-band 0.5 deg PPI at 0915-0930 UTC 10 November 2011
(upper) and vertical RHIs along the yellow line of
reflectivity (lower left) and hydrometeor identification
(lower right). |
|
||
|
|
|
| Figure 13. S-PolKa
S-band 0.5 deg PPI at 0950-1000 UTC 10 November 2011
(upper) and vertical RHIs along the yellow line of
reflectivity (lower left) and hydrometeor identification
(lower right). |
| The belt of easterlies at 200 hPa
continues to lie to the north of the "four-corners"
southern DYNAMO array (Figure 1).
The easterlies lie between generally broad anticyclonic
circulations to the north and south, however well defined
gyres local to the western Indian Ocean are not present in
either hemisphere at this time. Proceeding downward, the
wind direction in the array at 500 hPa has westerly
components, but is diffluent, with southwesterlies
prevailing at and north of the equator (i.e. the latitude
of Gan and the Revelle),
but is northwesterly over most of the array south of these
northernmost stations. At 850 hPa, winds were shearing
from southeasterly in the southern part of the array to
southwesterly in the northern part of the array. A similar
shearing was occurring at the 925 hPa, and the shearing at
that level extended over the S-PolKa site. These low-level
patterns indicate that the experimental region was lying
between major circulation features. These wind tendencies
seen in the maps of Figure
1 are consistent with the soundings at Gan and at
the Revelle (Figure 2). Both stations show the
light easterly component winds near the surface. The layer
of westerly component winds is much more pronounced at the
Revelle. The
METEOSAT infrared imagery shows Gan continuing to lie
under a long band of light convection extending far to the
southwest and northeast (Figure 3). This
band of convection lies in the shear zone of the 925 hPa
winds seen in Figure
1. In the vicinity of S-PolKa, cumulus congestus and
cumulonimbus produced local showers. Figure
4 shows several visual examples at various times
throughout the day. The layer clouds left behind at
different heights are seen in the two photos taken at 0857
UTC (middle right and lower left). Each of these showers,
and groups of showers, left behind cold pools evident as
holes on the S-PolKa S-band displays. An example of a
larger cold pool forming and expanding from a group of
showers near the radar is highlighted in Figure 5. Figure
6 shows the typical vertical structure of one of the
small convective cells. It was reaching about 8-9 km, and
the particle type algorithm (middle bottom) indicates
locally heavy rain, with graupel and larger nonmelting ice
particles (blue shading) in its upper portion. Humidity
layers are seen near the radar and some patchy layer cloud
is seen aloft in the RHIs. Similar structures can be seen
in the DOE Gan KAZR time-height series in Figure
7. Later in the day, some deeper convection,
reaching 14 km altitude, was seen at the outer southeast
limit of the radar range (Figure 8). |
|
|
|
|
| Figure 1.
Indian Meteorological Department 0-hour analyses for
0000 UTC 11 November 2011. |
 |
|
|
|
|
|
| Figure 2. DOE Gan
soundings for 11 November 2011. |
 |
|
|
|
|
|
| Figure 3. METEOSAT
infrared images for 11 November 2011. |
 |
|
|
| Figure 4. Photos
taken by Kaustav Chakravarty at the S-PolKa site at: Top
row, 0407 UTC 11 November 2011, facing east (left) and
northeast (right). Middle row: facing south, 0642 UTC
(left) and 0857 UTC (right). Bottom row: facing southeast
at 0857 UTC and northeast at 1040 UTC. |
 |
 |
|
| Figure 5. S-Polka 0.5
deg elevation PPIs for 11 November 2011 from 0207 to 0407
UTC. Boxes enclose region of a cold pool. |
|
||
|
|
|
| Figure 6. S-PolKa
S-band 0.5 deg PPI at 0315-0330 UTC 11 November 2011
(upper) and vertical RHIs along the yellow line of
reflectivity (lower left), hydrometeor identification
(lower middle), and radial velocity (lower right). |
 |
|
| Figure 7. Data from
the DOE Gan vertically pointing Ka-band KAZR radar
data for 11 November 2011. |
|
||
|
|
|
| Figure 8. S-PolKa
S-band 0.5 deg PPI at 1000-1015 UTC 11 November 2011
(upper) and vertical RHIs along the yellow line of
reflectivity (lower left), hydrometeor identification
(lower middle), and radial velocity (lower right). |
| Today we had deep convection and
precipitation in the area covered by S-PolKa. As we have
seen in previous cases, the convective cells moved with a
westerly component connected with low-level winds, while
the stratiform anvils produced by the deep convection were
sheared off to the west by upper-level easterlies. The
winds at 200 hPa were easterly and of low to moderate
intensity through the center of the four-corners array and
into the western Indian Ocean. These easterlies lay
between weak anticyclonic gyres north and south of the
equator (Figure 1, left). At the
850 hPa level, the winds across the northern tier of the
four-corners array were westerly and lying between
cyclonic circulations to the north and south (Figure 1, right). The DOE Gan 0600
UTC sounding also shows the easterly layer aloft and the
westerlies at lower levels (Figure 2).
Thus, as is common here, the convection was forming in a
sheared environment. The METEOSAT imagery for the day (Figure 3) showed Gan remaining
under the influence of the persistent SW-NE cloud band,
which was also present yesterday. A line of deep
convection that occurred within this band was approaching
S-PolKa at 0220 UTC 12 November 2011 (Figure
4). It was preceded by a gust front marked by a line
of enhanced reflectivity (upper left panel). This fineline
echo was marked by positive ZDR (red) in the upper right
panel. The lower panels show a convective cell in the line
with a top at about 14 km (lower left panel). The cell was
producing graupel and heavy rain and lofting large
nonmelting ice particles to about 12 km (middle panel).
The radial velocity showed a strong divergence signature
at echo top. A little over an hour later, a cell in this
same line had similar properties and was reaching an even
higher to a top height of 16 km (Figure
5). Figure 6 shows the
overal frequency of occurrence of convective and
stratiform echo top heights. Convective echoes even had 20
dBZ reflectivities occasionally occurring above the 15 km
level. The particle type statistics for the whole day (Figure 7) show the occurrence of graupel
(sometimes flagged as "small hail") up to about the 8 km
level. In general, the long line of convection seen in the
satellite imagery (Figure 3) was
not highly electrified, but lightning was reported from
the region where these convective cells were occurring (Figure 8). After the line passed,
we focussed more on the stratiform component of the
system. Figure 9 shows the
pattern of convective and stratiform echoes. The
convective echoes NNE of the radar were those associated
with lightning in Figure 8. In
this storm, the stratiform portions were relatively weak
in terms of rainfall and produced only 28% of the total
rain seen by the radar (Figure 10).
As on 31 October (see that summary) the shear probably
made it difficult for the stratiform regions to remain
connected with the convection. Therefore it could not
become an well formed stratiform region. Figure 11
shows that the leading edge of the main convective portion
of the line near S-PolKa was highly sloped back toward the
west, consistent with the shear. The small cell seen out
in front of the sloping cell was apparently a new cell
being triggered at the gust front seen in Figure 4. We had a good view of
this weak stratiform region shearing off to the west of
its convective source as the convective portion of
the system passed over S-PolKa as it moved towards the
southeast (Figure 11, right panel). Figure 12 shows the stratiform echo
becoming increasingly separated from the eastward
propagating convective elements. The sequence of vertical
cross sections in Figure 13 shows the
weakening structure of the stratiform precipitation region
as it moves off to the southeast. It was never a really
strong stratiform structure but at the early time in the
figure it has a moderately well defined structure with
melting aggregates ("wet snow") and some graupel
signatures above and below the melting layer. The graupel
signatures disappear, and the wet snow diminishes with
each successive time. The photographs in Figure 14 show the underside of the
stratiform anvil. The cloud line seen in the left panel
may have been on the edge of the cool wake of the system.
At the time of the photographs, rain was occurring at Gan,
where the DOE KAZR is located. Close inspection of the
middle photograph shows the rain. That rain was occurring
on the edge of the anvil cloud, and the KAZR data at about
0600 UTC in Figure 15 shows the rain.
After that time the KAZR data show the stratiform anvil
cloud with fall streaks. |
|
|
| Figure 1.
Indian Meteorological Department 0-hour analyses for
0000 UTC 12 November 2011. |
| Figure 2.DOE
Gan sounding for 0600 UTC 12 November 2011. |
 |
|
| Figure 3. METEOSAT
infrared (top) and visible (bottom) images for 0400 UTC 12
November 2011. |
 |
||
 |
 |
 |
| Figure 4. S-PolKa
S-band radar data for 0220-0232 UTC 12 November 2011.
Upper panels show PPIs of reflectivity (left) and
differential reflectivity (right) at 0.5 deg elevation.
Lower panels show reflectivity (left), particle type
(middle), and radial velocity (right) along the yellow
lines in the PPIs. |
|
||
|
|
|
| Figure 5. S-PolKa
S-band radar data for 0345-0354 UTC 12 November 2011.
Upper panels show PPIs of reflectivity (left) and
hydrometeor type (right) at 0.5 deg elevation. Lower
panels show reflectivity (left), particle type (middle),
and radial velocity (right) along the yellow lines in the
PPIs. |
| Figure 6. Frequency of
occurrence of echo tops at different reflectivity
thresholds as seen by the S-PolKa S-band radar on on 12
November 2011. |
| Figure 7.
Polarimetrically determined hydrometeor types dectected by
the S-PolKa S-band radar on 12 November 2011. |
| Figure 8. Lightning
flashes from the World Wide Lightning Location Network are
superimposed on the METEOSAT infrared image for 0430 UTC
12 November 2011. Locations of flashes occurring over 30
min prior to the satellite image are indicated by the red
and brown circles. |
 |
 |
| Figure 9. Convective
and stratiform echoes seen on the S-PolKa S-band radar at
the 2.5 km level on 12 November 2011. Yellow indicates
convective. Red indicates stratiform. |
| Figure 10. Convective,
stratiform and total rain seen by the S-PolKa S-band radar
by rain amount (upper) and rain area (lower) for 12
November 2011. |
|
|
 |
|
| Figure 11. S-PolKa
S-band reflectivity at 0317 (left) and 0402 (right) UTC 12
November 2011. Upper panels show interpolated reflectivity
in the RHI sectors of S-PolKa superimposed on the METEOSAT
infrared image. The lower panels show vertical cross
sections taken NW to SE along the red lines in the upper
panels. |
 |
 |
|
| Figure 12. S-PolKa
S-band radar PPIs of reflectivity (left) at 0.5 deg
elevation for 0546 (left), 0616 (middle), and 0646 (right)
UTC 12 November 2011. |
 |
 |
|
|
|
|
| Figure 13. S-Polka
S-band observations at 30 min intervals. Left panels show
PPIs of reflectivity. Right panels show RHIs of
hydrometeor type. |
|
|
|
| Figure 14. Photos
taken by Kaustav Chakravarty at the S-PolKa site, facing
NE (left), E (middle), and SE (right) at 0551 UTC 12
November 2011. Location of Gan is indicated for
reference to the KAZR data. |
 |
|
| Figure 15. Data from
the DOE Gan vertically pointing Ka-band KAZR radar
data for 12 November 2011. |
| Today featured a wide spectrum of
events on radar, from boundary layer echoes to
nonprecipitating cumulus to isolated storms with some
electrified. The wind patterns in Figure
1 show that the large-scale flow was easterly across
the four-corners array at high levels (200 hPa, upper left
panel). At 700 hPa the conditions were moist across the
northern tier of the array, corresponding to the region of
scattered high cloudiness that has been occurring along
this belt stretching from SW to NE of the array (upper
right panel). At 850 hPa, the winds were light but
westerly in the northern part of the array and easterly in
the southern portion of the array (lower left panel). The
winds were light westerly across the array at 925 hPa
(lower right panel). The DOE Gan soundings in Figure 2 show the moist layer
extending up to 500 hPa, coinciding with the westerlies in
the lower half of the troposphere. The sounding also shows
dry air at higher levels in the easterly layer. This
configuration of winds and humidity apparently provided a
ripe environment for convection, however it was not too
deep and lacked upscale development of mesoscale
organization. The upper portions of the convection were
encountering dry air and were sheared off into the
easterlies--as was the case yesterday (see yesterday's
summary). The infrared satellite data in Figure
3 showed upper level clouds in the vicinity of Gan
all day, but they did not have cold tops and tended to be
isolated and cellular. The visible satellite pictures
during the day also showed isolated convective clouds
scattered over the region around Gan (Figure
4). Nonetheless a host of interesting phenomena were
observed by the S-PolKa S-band radar. Before discussing the clouds, we note an interesting boundary layer echo seen in the morning. The upper panel of Figure 5 shows the ZDR pattern close to the radar. Bright red positive ZDR signals were appearing just to the inside of two of the islands on the southeast side of the atoll. According to the Equator Village dive boat operator, that region was experiencing waves crashing on the outside of the reef with huge plumes of sea spray that looked like "smoke" from 20 km away. He said the winds would have been carrying the spray to the inside of the reef. The positive ZDR signal was coming exactly from this region of breaking waves and spray. In the lower panels of Figure 5, the positive ZDR signal extends nominally to 300 m, although this height may be artificial if the rays were bending. The echo is also present in the reflectivity field in the lower right panel. This observation needs explanation but may be a clue for other positive ZDR signals we are seeing connected with the sea surface and finelines. In the mid-to-late morning, numerous boundary layer cumulus clouds were present, some of these clouds grew into congestus and small cumulonimbus. This evolution is illustrated by the photos in the upper row of Figure 6, which shows photos taken at three successively later times as the day progressed. As the showers dissipated, the remains of the clouds spread out laterally, as illustrated by the photos in the lower row of Figure 6. On the S-PolKa S-band radar, the nonprecipitating clouds in the boundary layer in midmorning were appearing in lines oriented WNW-ESE in reflectivity (upper left of Figure 7), parallel to the low-level winds (seen to be WNW in the radial velocity PPIin the upper right panel of Figure 7). The RHIs in the lower panels show humidity gradient layers lying above the cumulus clouds. The rhoHV shows the red positive rhoHV values in the mantle echoes surronding the clouds. The ZDR values are zero in the mantle echoes. As we have noted in previous summaries, both the mantle echoes and the humidity layers are thought to be produced by Bragg scattering at strong gradients of refractive index. As the day wore on, convective showers formed out of the cloud lines and subsequently continued to form on cold pool boundaries and intersections of cold pool boundaries. Figure 8 shows examples of the PPI patterns in the early afternoon. The convective cells can be seen along and at intersections of cold pools, which are the bare holes in the echo pattern. Figure 9 shows examples of the vertical structures of showers seen at the time of the PPIs in Figure 8. They were reaching 10-12 km and producing moderate amounts of ice aloft. In the early stage of development, seen in the upper row, some graupel was being produced. During later stages the convection was turning into a small stratiform structure (lower row). In the evening, a strong cell located along a large cold pool boundary passed over Gan, and we heard thunder at 1551 UTC. The PPIs in the upper row of Figure 10 show that the strongest cell was at the leading edge of a cold pool boundary. The cold pool boundary was marked by the typical (but as yet unexplained) positive (red) ZDR values. In view of the discussion of Figure 5, we wonder if the gust front produces white caps and spray that enhance the sea clutter signal; but this remains a question. Some echoes were reaching 14 km and were producing graupel (see bottom row, 3rd from left, in Figure 10). A small hail signature was present at about the 5.5 km level. This time corresponds to the audible thunder and visible lightning at Gan. The radial velocity panel shows the gust front in the low-level outbound velocities on the southeast side of the echo. Infrared satellite data showed a cloud top temperature of about 210 deg K (Figure 11). |
|
|
|
|
| Figure 1.
Indian Meteorological Department 0-hour analyses for
0000 UTC 13 November 2011. |
|
|
| Figure 2. DOE
Gan sounding for 0600 and 1200 UTC 13 November 2011. |
 |
 |
 |
| Figure 3. METEOSAT
infrared images for 13 November 2011. |
 |
|
| Figure 4. METEOSAT
visible images for 13 November 2011. |
|
||
|
|
|
| Figure 5. S-PolKa
S-band radar data for 0445-0500 UTC 13 November 2011.
Upper panel shows PPI of ZDR at 0.5 deg elevation. Lower
panels show ZDR (left) and reflectivity (right) along the
yellow line in the PPI. Range rings in the PPI are at 25
km intervals. |
|
|
|
|
|
|||
| Figure 6. Photos
taken by Kaustav Chakravarty at the S-PolKa site on 13
November 2011. Upper row, left to right: facing NE at 0600
UTC, E at 0703 and 0916 UTC. Lower row, left to right: S
at 0918 UTC and SE at 1052 UTC. |
 |
||
 |
 |
 |
| Figure 7. S-PolKa
S-band radar data for 0600-0615 UTC 13 November 2011.
Upper panels show PPIs of reflectivity (left) and radial
velocity (right) at 0.5 deg elevation. Lower panels show
RHIs of reflectivity (left), rhoHV (middle), and radial
velocity (right) along the yellow line in the PPIs. |
 |
|
| Figure 8. S-PolKa
S-band PPIs on 13 November 2011 at 0916 and 1001 UTC. |
 |
 |
|
|
| Figure 9. RHIs of
radar echo structure seen by the S-PolKa S-band radar at
around 1000 UTC on 13 November 2011. Reflectivity on the
left and particle type on the right. |
 |
|
|
 |
 |
|
|
|
| Figure 10. S-PolKa
S-band radar data fields at 1530-1545 UTC 13 November
2011. Left to right: reflectivity, ZDR, hydrometeor type,
and radial velocity. |
| Figure 11. METEOSAT
infrared image for 1600 UTC 13 November 2011. |
| The synoptic maps in Figure 1 show light winds at all
levels in the vicinity of the S-PolKa radar (0.7 S, 73.2
E). The soundings at Gan are consistent with this picture,
but they show an easterly wind direction from about 400
hPa upwards (Figure 2). The most
notable aspect of the soundings is that the layer below
400 hPa is moderately moist, while the upper levels are
very dry. Satellite data show sparse high cloudiness near
Gan, with most deep and wide convective systems far to the
south (Figure 3). The cloud
top temperatures are mostly warmer than ~225 K in the
vicinity of Gan. Precipitating convection nevertheless
occurred over the region surrounding Gan, mostly in
association with cold pool triggering. An example of a
cell that formed along a cold-pool boundary is shown in Figure 4. The RHI panels show
that the cell barely reach high enough to produce ice and
then collapsed. It reached only about 7 km. Perhaps it
could not penetrate the dry air above that level shown in
the soundings in Figure
2. Although the shear was relatively weak, these
small convective cells forming at cold pool boundaries
were also struggling against the shear, as illustrated by
the cell shown in Figure 5,
whose top was being sheared off--note the inbound easterly
velocity components in Figure
5. Very heavy rain occurred at the radar site when
several cold pool boundaries collided directly over Addu
Atoll. The sequence of PPIs in Figure 6 shows cold
pools coming from the northeast and southwest and
southeast and colliding at about 0515 UTC (upper right
panel). The convection that formed was very intense. Figure 7 shows the most intense
cell that developed. It had heavy rain with echoes ~55 dBZ
(middle row, left panel), signatures of graupel above and
below the 0 deg C level (middle row, center panel), and a
strong gust front near the surface in the radial velocity
(middle row, right panel). However, again we note that the
echo top did not extend to high levels. It reached only
~11 km. The infrared temperature showed only one pixel at
220 K, which was the coldest top. The combination of shear
and dry conditions aloft was probably preventing these
strongly convective elements from extending to high
levels. As a result these clouds could not organize
upscale and form large stratiform regions. In late
afternoon, the S-PolKa site was located within a cold pool
bounded by lines of clouds and high ZDR (Figure 8). A panorama of the cloud
line in the NW-N-NE-E sector visible from the S-PolKa site
is shown in Figure 9. |
|
|
|
|
| Figure 1. Indian Meteorological Department 0-hour analyses for 0000 UTC 14 November 2011. |
|
|
|
 |
| Figure 2. DOE Gan sounding for 14 November 2011. |
 |
|
|
|
|
|
| Figure 3. METEOSAT images for 0800 UTC 14 November 2011. Left: Infrared temperature with different color schemes. Right Visible image at different magnifications. |
|
||
|
|
|
| Figure 4. S-PolKa
S-band radar data for 0345-0425 UTC 14 November 2011.
Upper panel shows PPI of reflectivity at 0.5 deg
elevation. Lower panels show RHIs of hydrometeor type.
Range rings in the PPI are at 25 km intervals. |
|
|
| Figure 5. S-PolKa
S-band radar data for 0345 and 0414 UTC 14 November 2011.
PPI (left) shows reflectivity at 0.5 deg elevation. Right
panels shows an RHI along the yellow line of hydrometeor
type. Range rings in the PPI are at 25 km intervals. |
 |
|
|
|
|
|
| Figure 6. S-Polka S-band 0.5 deg elevation PPIs for 14 November 2011. |
|
||
|
|
|
|
||
| Figure 7. S-PolKa
S-band radar data for 0550-00600 UTC 14 November 2011 (top
and middle rows). RHIs in the middle row show dBZ,
hydrometeor type, and radial velocity. Lower panel shows
the 0600 UTC infrared temperature from
METEOSAT. Range rings in the PPI are at 25 km
intervals. |
 |
|
|
| Figure 8. S-Polka
S-band PPIs at 0.5 deg elevation at 1116 UTC 14 November
2011. |
| Figure 9. Panoramic
photos taken by Hannah Barnes at the S-PolKa S-band site
at 1115 UTC 14 November 2011. The jetty points east. |
| Today
featured significantly less convective activity in the
S-PolKa domain than the previous two days. While the
200 hPa synoptic map (left panel
of Figure 1) shows that
the near equatorial region near Gan and the Revelle have
southeasterlies, most of the southern DYNAMO array has
easterlies. Conditions at 700 hPa (center panel of Figure 1) are
characterized by light and variable winds near Gan and
the Revelle
and easterlies near Diego Garcia and the Mirai. At 925
hPa (right panel of Figure 1) the entire
southern DYNAMO array is characterized by
easterlies.Figure 2 shows that
soundings taken in Male, Gan, Diego Garcia, and at the
Revelle at
0601 UTC have light winds through 200 hPa. While Diego
Garcia and the Revelle have an easterly component
through the depth of the troposphere, Male and Gan
have a small region of westerlies at approximately 750
hPa (2 - 3 km altitude). This wind profile is
consistent with the maps discussed
in Figure 1.
Additionally, each of the soundings have dry
conditions above 600 hPa. The upper levels have
exhibited a drying trend for the last few days in
association with large-scale
subsidence. Figure 3 shows a
large maximum of positive 200 hPa velocity potential
over the central Indian Ocean. Positive upper level
velocity potential is indicative of convergence and
sinking motion through the column. Evidence of
large-scale subsidence is also seen in the soundings.
Each of the soundings in Figure 2
have a slight subsidence inversion at approximately 550
- 500 hPa. Brandon Kerns also notes this large-scale
subsidence in the 14 November 1800 UTC
Daily Weather Briefing. The most significant convection today was west of the DYNAMO array. Figure 4 (top) indicates that a large amount of lightning occurred in the half hour preceding 0531 UTC. This high flash frequency was observed throughout the day in these two regions. The bottom panel of Figure 4 indicates that the infrared temperatures at 0531 UTC in the three electrically active convective centers were near 200 K. Within the S-PolKa domain, the day started with a number of large cold pools. Not only were these cold pools evident in the reflectivity, but they were also associated with large differential reflectivity values, which has been a common occurrence throughout DYNAMO (top panels of Figure 5). Vertical cross sections through these cold pools indicate that these features extended to approximately 800 m in altitude. The differential reflectivity (upper right panel of Figure 5) had exceptional high values near islands, especially on the western side of Addu Atoll. These are some of the highest differential reflectivities we have observed and it is unclear why this enhancement is occurring. The top panel of Figure 4 shows that the belt of convection that has been affecting the S-PolKa domain for the last few days has moved to the south. Thus, S-PolKa was located between the strong convection to the northwest and the belt of larger cumulus to the south. During the late afternoon shallow cumulus and cumulonimbus began to form in northwest - southeast oriented lines as illustrated in the top panel of Figure 6. Infrared temperatures at 1101 UTC (bottom panel of Figure 6) indicate that the temperatures were only about 260 K. At later times, cloud top temperatures reached a minimum of 250 K. These shallow lines of cumulus, when viewed in time lapse, were oriented and moved from northeast to southwest. This motion is consistent with northeasterly flow observed at low levels throughout the day in the Gan soundings (upper right panel of Figure 2). A representative example of afternoon convection observed in the S-PolKa domain today is shown in Figure 7. Even the strongest cells barely reached 8 km and quickly dissipated. The shallow band of westerlies noted in the Gan sounding above (upper right panel of Figure 2) is evident in Figure 7 as a band of outbound velocities at approximately 3 km. Figure 8 visually shows the variation of the cloud population throughout the day. Morning convection was limited to shallow cumulus (left panel). However, by early afternoon deeper cumulus began to form in lines (center panel) and isolated cumulonimbus developed in the late afternoon (right panel). Additionally, the left panel of Figure 8 shows that the anvils were shearing towards the east, which is uncommon. |
 |
 |
 |
| Figure 1.
Indian Meteorological Department 0-hour analyses for
0000 UTC 15 November 2011. |
 |
 |
|
|
| Figure 2.
Soundings at Male (upper left), Gan, (upper right), the
Revelle (lower
left), and Diego Garcia (lower left) at 0601 UTC 15
November 2011. |
| Figure 3. 200
hPa Velocity Potential anomolies (positive brown
contours, negative green contours) and daily IR
temperature(fill) for 14 November 2011 from the CPC MJO
working group. |
|
 |
| Figure 4. Top
Panel: Lightning flashes from the World Wide Lightning
Location Network superimposed on METEOSAT infrared
imagery at 0531 UTC 15 November. All flashes over the 30
minutes prior to 0531 UTC are shown. Bottom Panel:
METEOSAT infrared images at 0531 UTC 15 November 2011 on
a different color scale. |
|
|
|
|
| Figure 5. Top
Row: SPolKa S-band 0.5 degree elevation PPIs for 0046 -
0501 UTC 15 November 2011 reflectivity (left panel) and
differential reflectivity (right panel). Bottom row:
RHIs along the yellow line shown in the PPIs above of
reflectivity (left panel) and differential reflectivity
(right panel). |
|
|
| Figure 6. Top Panel:
METEOSAT visible imagery. Bottom Panel: METEOSAT infrared
imagery at 1101 UTC 15 November 2011. |
 |
|||
 |
 |
 |
| Figure 7. SPolKa
S-band radar data valid from 1031 - 1046 UTC 15 November
2011. Top Panel: PPI of reflectivity. Bottom Panel: RHIs
along the yellow line in the PPI of reflectivity (left
panel), hydrometeor type (center panel), and radial
velocity (right panel). |
|
 |
|
| Figure 8. Photos
taken by Kaustav Chakravarty at SPolKa site on 15 November
2011. Right Panel: 0500 UTC facing south. Center Panel:
0941 UTC facing north-northeast. Left Panel: 1038 UTC
facing north-northwest. |
| Today Diego Garcia experienced the
development of a convective ring of the type that has been
observed numerous times throughout DYNAMO. Meanwhile at
the S-PolKa sight we observed a spectacular display of
cold pools. The NOAA P-3 aircraft flew a mission that
sampled both the convective outbreak and the cold
pools. The winds were exceptionally weak at all levels
in the vicinity of S-PolKa, as can be seen from the maps
in Figure 1. The DOE Gan
soundings in Figure 2 show
that a coherent layer of weak easterlies layer extended
from 600 to 250 hPa, but at 200 hPa the winds were light
and variable. At 850 hPa, the winds were northeasterly
in the vicinity of S-PolKa, and convective echoes were
translating in a NE to SW direction. The soundings also
show a richly moist layerup to 600 hPa and dry air
above. The soundings all show a subsidence-type stable
layer just above 600 hPa, where the air becomes dry. The
subsidence is consistent with the upper level velocity
potential discussed in yesterday's summary. Figure 3 illustrates convective outbreak and expanding convective ring emanating from near Diego Garcia. The figure shows the METEOSAT infrared temperatures every three hours starting 1800 UTC 15 November until 1200 UTC 16 November 2011. The convection that generated ring developed overnight within a band of deep convection oriented southwest to northeast through Diego Garcia. By 0600 UTC 16 November 2011 Diego Garcia is located within the clearing in the center of the expanding ring of convection. The upper left-hand panel of Figure 4 illustrates that lightning was prevalent during the initial burst of convection at 2100 UTC 15 November. However, as the convective ring expanded the flash frequency diminished (upper right hand panel of Figure 4). The bottom row of Figure 4 shows the corresponding METEOSAT infrared imagery in a different color scale. During both time periods lightning appears to be located in storms with infrared temperatures near 200 K. In the remainder of this summary, we discuss the cold pools that were occurring near S-PolKa from the evening of the 15th and continuing through the day on 16 November. Figure 5 shows the sequence of cold pool echoes that occurred over night. Note how they cold pool boundaries are again marked by enhanced positive ZDR, seen as red in the images. Figure 6 shows some details of one of the cells that formed at the edge of one of the nocturnal cold pools at 2246 UTC 15 November (see also left panels of Figure 5). The cell was only reaching 9 km but was very intense. The polarimetric hydrometeor type algorithm (middle bottom) was flagging "hail" at one pixel--although it was probably just large or concentrated graupel. The radial velocity was showing a classic updraft downdraft pair wih a strong inbound gust front (bottom right of Figure 6). Figure 7 shows the structure of another cell forming at the boundary of the cold pool seen NE of the radar at 0501 UTC 16 November (between the times of the middle and right panels of Figure 5). The cell in Figure 7 was reaching 15 km altitude and producing a small anvil. It also contained graupel (bottom middle panel of Figure 7) and an updraft-downdraft pair, gust front, and echo-top divergence in the radial velocity (bottom right panel of Figure 7). Figure 8 shows the anvil of the cell seen at the cold pool boundary about 60 km NE of S-PolKa. The anvil looked somewhat confused as to which way it was streaming off in the light winds aloft (left photo). As it developed and separated from its parent cumulonimbus is exhibited mammatus (middle and right panels). Figure 9 shows the development of the cold pool sampled by the NOAA P3 aircraft. At 0831 UTC 16 November a small rain cell was located on a cold pool boundary about 45 km ENE of S-PolKa. A cold pool was seen forming around the rain area at 0931 UTC. The cold pool expanded, and at 1001 UTC another patch of rain cells was located about 70 km ESE of the radar. By 1031, the second rain area was form its own cold pool, and the rain in the center of the earlier cold pool was weakening. This process continued through 1101 UTC. By 1131 UTC, the origina rain cell had dissappeared and only its wide cold pool remained and was intersecting the cold pool of the second rain area. Figure 10 shows a set of cross sections throug the second rain area. It is the rightmost convective element in the figure, and it can be seen to have been reaching 12 km and producing a moderately thick anvil of limited horizontal extent. Figure 11 shows photographs of the first cell to the southeast. |
 |
 |
|
| Figure 1.
Indian Meteorological Department 0-hour analyses for
0000 UTC 16 November 2011. The region around Gan is
circled. |
 |
|
|
| Figure 2.
Soundings at Gan for 16 November 2011. |
|
|||
|
|
|
 |
| Figure 3.
METEOSAT infrared images every three hours from for 16
November 2011. |
 |
 |
 |
 |
| Figure 4.
METEOSAT infrared satellite imagery for 16 November
2011. Top panels show 30 min accumulated lightning
flashes from the World Wide Lightning Location Network. |
 |
 |
|
 |
|
|
|
Figure 5.
S-PolKa S-band 0.5 deg elevation PPIs 2246 UTC 15
November 2011 and 0401 and 0701 UTC 16 November 2011.
Reflectivity in upper row. Differential reflectivity in
lower row. |
 |
||
 |
|
|
| Figure 6.
S-PolKa S-band radar data for about 2240 UTC 15 November
2011. Left to right: upper row: dBZ, ZDR; lower row:
dBZ, hydrometeor type, radial velocity. |
 |
||
 |
|
|
| Figure 7.
S-PolKa S-band radar data for about 0445-0500 UTC 16
November 2011. Upper row: dBZ. Lower row, left to
right: dBZ, hydrometeor type, radial velocity. |
 |
 |
|
| Figure 8.
Photos taken looking NE from S-PolKa site on 16 November
2011 at 0655 UTC (left) and 0732 UTC. |
|
|
|
|
|
|
|
||
| Figure 9.S-PolKa
S-band 0.5 deg elevation PPI sequence for
0831-1101 UTC 16 November 2011. Reflectivity in upper
row. Differential reflectivity in lower row. |
 |
||
 |
|
 |
| Figure 10.
S-PolKa S-band radar data for 1130-1145 UTC 16 November
2011. Upper row: dBZ. Lower row, left to right:
dBZ, hydrometeor type, radial velocity. |
 |
 |
|
| Figure 11.
Photos taken looking ESE from S-PolKa site on 16
November 2011 at 1225 UTC. Left to right, the photos
show three views of the same cloud. |
| Today marked a switch back to
low-level westerlies (Figure
1). By mid-day the layer of westerlies was up to 700
mb, which is a distinctive switch from the full column of
easterlies seen yesterday. At 200 mb the winds remain
easterly and relatively weak (Figure 2).
The 0000 UTC synoptic map indicates that the 700 mb winds
were weak and had a westerly component near S-PolKa, which
is indicated by the red dot. At 925 mb, the winds
had an easterly component. The layer of dry air beginning
at 600 mb overnight (2336 UTC 16 November) was down to 700
mb by midday, 0535 UTC 17 November
(Figure 3). The extensive
cirrus overhead in the morning cleared rapidly in the
late morning to the south and east of the radar, with
cloud tops rapidly evaporating over a matter of minutes
(Figure 4). There was enhanced convection to our north and south overnight, but a lack of convection over S-PolKa. The series of satellite images in Figure 5 show that convective lines oriented NW-SE and SW-NE grew and began to merge east of S-PolKa. However, most of the day lacked convection directly over S-PolKa. At the time of the first S-PloKa image shown here, the NW-SE line had moved north and an intense convective cell moved very close to the radar (Figure 6). A cross-section through the cell (Figure 7) shows the anvil reaching ~14 km in altitude (in this view) with a strong divergence signal at echo top, and a sloping updraft/downdraft couplet ~18 km from the radar. While the closer/older cell contained graupel below the melting level and a large quantity of aggregates, the farther/younger cell had graupel above and below the melting level and fewer aggregates. After the passage of this cell, convection was mostly restricted to the northern portion of the S-PolKa range, although convection began to wrap around from both the north and south over the course of the day (Figure 8). Convection in the two lines was electrified throughout the day, especially in the northern line (Figure 9). Anvil and cirrus continued to stream over S-PolKa throughout much of the day from the northern convective line (Figure 10). Time lapse movies of the radar data show that although the convective cell propagation was mostly westward yesterday, convective cell propagation shifted from westward in the morning to eastward midday and northward by the late afternoon today. The anvils also spread towards the west in the morning, but spread towards the east and south in the afternoon (Figure 11). Compared to the large prevalence of strong cold pools over the past two days, the convective cells had relatively few well-defined cold pools in the radar imagery. An example of one of these cold pools is shown in Figure 12. It is outlined by a weak echo cloud line in the left panel and by positive ZDR (red) in the right panel. As the convective cells began to wrap around S-PolKa, the amount of convective cells began to increase towards the east and south of the radar. By sunset, clouds were building to the south and west over the atoll and rained over several of the islands, which continued throughout the night (Figure 13). |
 |
| Figure 1 Time series of sounding winds
and relative humidity for the week of 11 Nov to 18 Nov. |
 |
 |
 |
| Figure 2 200 hPa, 700 hPa, and 925 hPa
IMD ARW 0000 UTC 17 November analyses. |
|
|
 |
| Figure 3.
Soundings at Gan for 17 November 2011. |
|
 |
 |
| Figure 4 Photos taken towards the SE of
SPolKa at 0536 UTC, 0705 UTC, and 0830 UTC. |
 |
 |
 |
 |
| Figure 5. Infrared satellite images for
0000-9000 UTC 17 November. |
 |
|
|
| Figure 6 SPolKa reflectivity PPI at 0153
UTC, 0206 UTC and 0216 UTC. |
 |
 |
 |
| Figure 7. 0223 UTC SPolKa reflectivity,
velocity, and particle identification RHI at 40 degrees
azimuth, shown in the right-most panel of Figure. |
 |
 |
 |
| Figure 8. Water vapor imagery for 0100
UTC, 0400 UTC, and 0700 UTC. |
 |
 |
| Figure 9 Visible satellite imagery with
WWLN flashes overlaid at 0000 UTC and 0600 UTC. |
 |
 |
| Figure 10 SPolKa reflectivity PPI at 0703
UTC and RHI at 54 degrees azimuth at 0710 UTC. |
 |
 |
| Figure 11 Photos taken towards the NE at
0536 UTC and 0947 UTC. |
 |
 |
| Figure 12 SPolKa reflectivity and ZDR PPI
at 0552 UTC. |
 |
| Figure 13 Photo taken towards the south
at 1253 UTC. |
| A major rain event occured
overnight 17-18 November and continued through 18
November. To summarize, the event began with the buildup
of convection mentioned at the end of yesterday's summary.
At 2300 UTC, the infrared satellite identified a very cold
cloud top directly on top of S-PolKa (upper-right panel of
Figure 1), which corresponded to
deep convection and high reflectivities (Figure 2). From 0300-0500 UTC, the
convection began to spread outward into a ring (upper
panels of Figure
3). As the western portion of the ring propagated outward, a
gust front was evident in the form of a roll cloud to the
west of the atoll around 0400 UTC (Figure 4). The convection
divided into lines north and south of S-PolKa beginning
around 0600 UTC (lower panels of Figure 3). The northern line
was more strongly electrified than the southern line (Figure 5). A line of
convection, which formed east of the radar, was initially
highly convective at 0800 UTC (Figure
6). This convection then collapsed into stratiform
by 1300 UTC (Figure 7).
Tendrils of convection formed and decayed northeast of the
line with no apparent preferred orientation (rightmost
panel of Figure 6). Heavy
rain at S-PolKa obscured the ability to take pictures most
of the day, but the heavy rain in the line to the east was
visible in the distance around 0600 UTC (Figure 8). The soundings show that the dry air reaching down to 650 hPa during the early evening yesterday raised to 550 hPa by 2030 UTC, 450 hPa by 0230 UTC, and 350 hPA by 0530 UTC (Figure 9). The winds near the surface are weakly westerly and the winds aloft remain weakly easterly. The model analyses show a weak low-level cyclonic gyre between 700-925 hPa that is directing air from the southern tip of India towards the region of the northern convective line (Figure 10). The remainder of the summary will highlight the convective episode over and to the south of S-PolKa overnight and during the early morning hours of 18 November 2011. During this period, convective regions mainly propagated towards the northeast while stratiform regions moved towards the northwest (Figure 11), as we have seen in previous cases and by previous investigators (see our summary for 26 October). Figure 12 shows the progression of convective and stratiform precipitation as it passed over S-PolKa between 2200 UTC 17 November and 0200 UTC 18 November. Initially the mid-level inflow (green shading between 5 and 12 km) was tilted upward toward the convection, but later in the period the inflow switched to being tilted downward, as the system became more stratiform (Figure 13). Most of the stratiform in the region showed very distinctive brightbands that sometimes reached to 50 dBZ with large quantities of melting aggregates along the melting level and non-melting aggregates above that (Figure 14). This period was also marked by interesting interactions between the convective cells and the anvils of stratiform regions. As lines of convection moved towards the leading edge of a stratiform region, the leading anvil seemed to wrap around the top of the convective cell before the cell was absorbed. (Figure 15). An example of this leading anvil interaction was also seen in the KAZR data (Figure 2). Similar to what was seen on 24 October, an "onion"-type sounding appeared towards the end of the day (Figure 16). During this time period, the region was predominantly covered in stratiform precipitation. |
 |
 |
|
 |
 |
 |
| Figure 1. Infrared satellite imagery from
2200 UTC 17 November - 0200 UTC 18 November. |
 |
| Figure 2 KAZR
co-polarized reflectivity data from 0000 UTC 18 November
to 0000 UTC 19 November. |
 |
 |
 |
 |
 |
 |
| Figure 3. Infrared
satellite imagery from 2200 UTC 17 November - 0200 UTC 18
November. |
 |
| Figure 4. Photo looking west from the
causeway between Gan and Feydhoo at 0413 UTC. |
 |
| Figure 5. Visible satellite imagery and
WWLN lightning flashes for 0700 UTC 18 November. |
 |
 |
 |
|
| Figure 6. SPolKa reflectivity PPI for
0816-1016 UTC 18 November. |
 |
 |
 |
 |
| Figure 7. SPolKa convective/stratiform
seperation for 1100-1300 UTC 18 November. |
 |
| Figure 8 Photo taken facing east at 0657
UTC. |
 |
 |
 |
 |
 |
 |
| Figure 9. Gan soundings 1736 UTC 17
November - 1136 UTC 18 November. |
 |
 |
 |
| Figure 10. Indian ARW model analyses
output for 0000 UTC 18 November at 200 hPa, 700 hPa and
925 hPa. The red circle indicates the location of the
radar. |
 |
 |
 |
| Figure 11. SPolKa convective/stratiform
seperation for 1830-1930 UTC 17 November. |
 |
 |
 |
 |
| Figure 12. SPolKa reflectivity PPI at 0.5
degree elevation every hour from 2200 UTC 17 November to
0100 UTC 18 November. The line indicates the RHI azimuth
angle for Figure 13. |
 |
 |
 |
 |
 |
 |
| Figure 13. SPolKa reflectivity and velocity
RHI at 141.9 degrees azimuth (same as line in Figure 12)
for 2200 UTC and 2300 UTC 17 November and 0100 UTC 18
November. |
 |
||
 |
 |
|
| Figure 14. SPolKa reflectivity PPI and RHI
of reflectivity and particle identification at 20 degrees
azimuth for 0137 UTC 18 November. |
 |
 |
 |
 |
 |
 |
| Figure 15. SPolKa PPI and RHI reflectivity
at 126 degrees azimuth for 1016-1046 UTC 18 November. |
 |
||
 |
 |
|
| Figure 16. Sounding and SPolKa PPI
reflectivity, rain rate, and convective/stratiform
seperation for 1800 UTC 18 November. |
| Today we returned to locally
suppressed conditions after the highly active period
yesterday (Figure 1). Most
of the convection in the area was west of Diego Garcia far
to the southwest of Gan (Figure 2).
From 0000-1200 UTC 19 November convection formed and
decayed along a long line oriented northeast to southwest
without much movement (Figure
2) and a line of convection within the range of the
Mirai moved in
around 2000 UTC 18 November
(Figure 3).
The Mirai
soundings indicate that the winds were mostly easterly
through the column, and high relative humidities extended
up to 550 hPa during the passage of the convective line (Figure 4). The model analyses show a zone of 500 hPa confluence near the equator at about 70 E, but winds at all levels remain weak (Figure 5). The 700 hPa and 925 hPa maps indicate convergence near the convective line sampled by the Mirai. The winds below 600 hPa at Gan were initially weak and westerly, but shifted to easterly by 0600 UTC (Figure 6). In contrast, the winds above 300 hPa shifted from westerly to southerly by 1435 UTC, but were easterly by 2100 UTC. Near S-PolKa, we witnessed a return of the cold pools in the early morning hours. These cold pools are bounded by very large reflectivities and highly distinctive ZDR fields (Figure 7). While the background ZDR is usually noisy and averaging negative to zero during the day and positive at night, the daytime of ZDR field was positive throughout today. Compared to the same time of day on 9 November, the ZDR is distinctly positive on 19 November (Figure 8). Additionally, while particle identification classified most of the echoes on 9 November as second trip (gray), many of the echoes on 19 November were classified as insects (white). Coincidentally, this time of year corresponds to the maximum in dragonfly migration across the Indian Ocean. The sky was very clear and bright blue, with a cloud population mostly characterized by shallow cumuli, thin mid-level clouds (Figure 9), and the occasional anvil from passing convective cells (Figure 10). In the afternoon the shallow clouds began to orient themselves along lines extending northeast to southwest (Figure 11), and eventually built into a relatively weak line of convective precipitation. |
 |
| Figure 1. Photo taken towards the east at
0652 UTC 19 November. |
 |
|
|
|
 |
 |
| Figure 2. Infrared satellite imagery
from 2100 UTC 18 November - 1200 UTC 19 November. |
 |
| Figure 3. Mirai radar
reflectivity PPI for 0000 UTC 18 November to 0000 19
November. |
 |
| Figure 4. Mirai sounding
weekly time series from 13-20 November. |
 |
 |
 |
 |
| Figure 5. IMD model analyses of 200 hPa,
500 hPa, 750 hPa, and 925 hPa for 0000 UTC 19 November. |
 |
 |
 |
 |
| Figure 6. Gan soundings every 6 hours from
2100 UTC 18 November - 2100 UTC 19 November. |
 |
 |
| Figure 7. SPolKa Reflectivity and ZDR PPI
for 2316 UTC 18 November. |
 |
 |
|
 |
 |
 |
| Figure 8. SPolKa reflectivity, ZDR, and PID
PPI for 0616 UTC 09 November (top row) and 19 November
(bottom row). |
 |
| Figure 9. Photo looking ENE at 0658 UTC. |
 |
 |
 |
| Figure 10. Photos looking SE at 0404
UTC, 0425 UTC, and 0651 UTC. |
 |
| Figure 11. S-PolKa reflectivity PPI at 0931
UTC 19 November. |
| As of yesterday, a zone of
negative velocity potential is building over Africa and
beginning to extend into the western Indian Ocean (Figure 1). This pattern is
indicative of upper-level divergence and rising motion
through the column. The central Indian Ocean is between
the region of enhanced
upper-level divergence (negative velocity potential) and
convergence (positive velocity potential). Upper-level
easterlies are increasing over the region as the northern
part of the array is under the difluence region of two 200
hPa anticyclones positioned north and south of the equator
(Figure 2). Low-level (925-850
hPa) westerlies east of the cyclonic rotation in the
southeastern Indian ocean are meeting the low-level
easterlies in the central Indian ocean, creating a slanted
line of convergence stretching roughly from 0-10 S, 70-75
E. This line of convergence was colocated with a line of
convection between Diego Garcia and Revelle (Figure 3). A burst of convection in
the middle of the array showed initial signs of outward
propagation, but the outward propagation was limited. The
convective line just south of India that has been
mentioned over the past few summaries remains highly
electrified, but the line of convection within the DYNAMO
array contained very little lightning (Figure 4) Within the range of S-PolKa, mainly shallow convection occurred, and it moved mostly westward throughout the day (Figure 5). Winds remained light and primarily easterly through the column, with the exception of a layer of weak westerlies below 900 hPa (Figure 6). Conditions were dry down to 700 hPa. During the late morning, smaller convective cells to the east of the radar merged to form a more intense cell (Figure 7 and Figure 8). Just after the merger, the cells intensified briefly (Figure 9), getting just high enough to produce a small amount of ice cloud that streamed westward. This pattern of merging cells was common in the morning. The two cells seen in the last image of Figure 8 merged to form the tallest cell seen during the day (Figure 10), but the height of the cell was only just enough to produce ice cloud and a small amount of graupel at its top. In general, all of the cold-cloud particle types were in much lower in frequency and in confined to much lower altitudes than the rain event on 18 November (Figure 11, note in the caption that this comparison is subject to the caveat that the radar was off for a substantial time period on 20 November). |
 |
| Figure 1. 200 hPa velocity potential
anomaly overlaid the daily infrared satellite imagery for
19 November. From NCEP
CPC. |
 |
 |
 |
| Figure 2. IMD
model analyses at 200, 850, and 925 hPa for 0000 UTC 20
November. The red box indicates the location of the DYNAMO
array. |
 |
 |
 |
 |
| Figure 3. Infrared
satellite imagery every three hours from 0000-0900 UTC 20
November. |
 |
| Figure 4. WWLN lightning data overlaid the
infrared satellite imagery for 0400 UTC 20 November. |
 |
 |
|
|
| Figure 5. Photos taken looking SE at 0427
UTC, 0549 UTC, 0828 UTC, and 1148 UTC 20 November. |
 |
 |
 |
| Figure 6. Gan soundings at 2335 UTC, 0236
UTC, and 0536 UTC 20 November. |
 |
 |
 |
 |
| Figure 7. Photos spanning NE to SE
at 0614 UTC 20 November. |
 |
 |
 |
 |
 |
 |
| Figure 8. S-PolKa reflectivity PPI every
15 minutes from 0346 UTC to 0501 UTC 20 November. The
yellow line indicates the RHIs
in Figure 9. |
 |
 |
 |
 |
 |
 |
 |
 |
| Figure 9. S-PolKa reflectivity and PID RHI
every 15 minutes from 0411 UTC to 0455 UTC 20 November
along the yellow line in Figure 8. |
 |
||
 |
 |
 |
| Figure 10. S-PolKa reflectivity PPI at
0631 UTC 20 November and reflectivity, PID and velocity
RHI along the yellow line at 0625 UTC 20 November. |
 |
 |
| Figure 11. S-PolKa PID statistics by
altitude for 18 and 20 November. The comparison of these
two figure panels is subject to the caveat that the
S-PolKa radar was taken offline for a technical problem
just after 0630 UTC 20 November. |
| Convection remains active across
much of the Indian Ocean (Figure 1),
with two distinct lines crossing most of the basin. The
line to the north remains highly electrified, and the line
to the south seems to be slowly approaching the Equator.
Diego Garcia experienced lightning with a cell that passed
over the island. A large area of convection corresponding to a large low-level cyclonic gyre remains mostly stationary in the southwestern Indian Ocean (Figure 2). The coherence of the northern hemisphere 200 hPa anticyclone has broken down somewhat on its eastern side, but the southern hemisphere gyre and strong easterlies along the equator remain. Negative velocity potential, corresponding to enhanced upper level divergence and rising motion throught the column, continues to build eastward from Africa (Figure 3). It now covers much of the western Indian Ocean and has a wavenumber one structure, which suggests that an MJO may initiate soon. The winds at S-PolKa have returned to being mostly westerly below 800 hPa (Figure 4). By 0600 UTC the layer between 600-800 hPa had a strong southerly component and the column had moistened up to 500 hPa. While moisture at the Revelle has been steadily climbing over the past week, the moisture signal at Gan is less clear (Figure 5). In both time series, a switch to low-level westerlies appears to roughly correspond with an increase in upper-level moisture, whereas deep-layer easterlies are roughly associated with dry upper-levels. In general, the echo tops measured by S-PolKa have been slowly increasing in height since the beginning of the month (Figure 6). By 18 November the 10 dBZ echo tops consistently reach over 15 km. In the 30 dBZ echo top statistics, there is an interesting cycle of roughly 3 days of increasing height followed by two days of suppressed heights. In general, the deep and shallow 30 dBZ echo tops have also been trending upwards. The morning was cloudy with occasional showers, but by late afternoon the showers had moved towards the northwest and the sky cleared except for the trailing anvil left behind by the morning's showers (Figure 7). One of the tallest convective cells seen during the day was part of a line that formed south of Gan and propagated towards the east and then north-northeast through the area between 0400-0800 UTC (Figure 8). This cell's echo reached about 13 km on the SMART-R C-band radar, with ~30 dBZ echo reaching 8-10 km. The stratiform cloud and rain from this cell moved north-northwest through the morning and passed directly over KAZR just after 0600 UTC (Figure 9). A brightband just below 5 km and portions of the storm's trailing anvil was captured by the radar. The high clouds seen throughout the day (Figure 10) are also evident in the KAZR reflectivity. Holes in the high cloud information directly above regions of heavy precipitation (~0700 and 2200 UTC) are likely the result of attenuation of the KAZR beam (Figure 9). Late in the day, the convection northwest of SMART-R switched to moving towards the east while its stratiform portion moved westward (Figure 11). A strong cell of precipitation passed over Gan overnight, which will be discussed in tomorrow's summary. |
 |
| Figure 1. WWLN lightning data overlaid the
infrared satellite imagery for 0600 UTC 21 November. |
 |
 |
 |
| Figure 2. IMD model analyses at 200 hPa,
700 hPa, and 925 hPa for 0000 UTC 21 November. The red box
indicates the location of the DYNAMO array. |
 |
| Figure 3. 200 hPa velocity potential
anomaly overlaid the daily infrared satellite imagery for
20 November. From NCEP
CPC. |
 |
|
 |
| Figure 4. Gan soundings for 0000, 0600,
and 1200 UTC 21 November. |
 |
| Figure 5. Sounding timeseries for Gan
(upper panel) and the Revelle
(lower panel) for the week of 14-22 November. |
 |
 |
| Figure 6. S-PolKa echo top statistics for
01-18 November. |
 |
 |
 |
 |
| Figure 7. Photos looking ESE at 0407 UTC,
0543 UTC, 0921 UTC, and 1127 UTC 21 November. |
 |
||
 |
 |
 |
| Figure 8. SMART-R reflectivity PPI at 0516
UTC, and reflectivity RHI along the 146 degree azimuth for
0420 UTC, 0510 UTC, and 0520 UTC. |
 |
| Figure 9. ARM KAZR co-polarized
reflectivity data for 21 November. |
 |
| Figure 10. Photo looking south at 0921 UTC
21 November. |
 |
| Figure 11. Photo looking north at 1128 UTC
21 November. |
| Today was a showery day at S-PolKa
and SMART-R as convection continues to build in the region
(Figure 1). Overnight convection
began around 1800 UTC 21 November and produced a
significant amount of rain over Gan. These convective
cells generally came from the northwest and were a bit
taller and more intense than those of the past few days,
with higher reflectivities, broader and deeper updrafts,
and echotops reaching about 9 km (Figure
2). Additionally, these clouds were tall enough to
have ice in their tops, which is a change from the warm
convection yesterday morning. However, these cells did not
produce much stratiform within the view of S-Polka (during
the time that it was operating) or SMART-R (Figure 3). These cells grew in depth
throughout the day, leaving copious amounts of anvil cloud
fragments (Figure 4 and Figure 5). In the afternoon, a
convective cell to the northwest released a gust front
that propagated towards the S-PolKa site (Figure 6, Figure 7 and Figure 8). This cell occurred in
the cone of blockage for SMART-R and while S-PolKa was not
operational. At 1800 UTC 21 November, conditions at Gan were dry above 500 hPa (Figure 9), but by 0000 UTC 22 November conditions had moistened aloft, and by 1200 UTC the sounding was moist throughout the column. Winds remained westerly below 700 hPa throughout the day. A region of 925 hPa convergence was occurring along the equator in the vicinity of Gan (Figure 10). A weakly cyclonic region is directly over the DYNAMO array at 700 hPa, and winds have intensified at 200 hPa. Elsewhere in the array, the NOAA P-3 aircraft flew into a cluster of convective cells just west of the Revelle (Figure 11). The Falcon flew into the remnants of the same cluster later that evening. The Mirai and Diego Garcia are under the influence of extremely dry air which has been steadily moving westward for the past few days (Figure 12) and is associated with a significant drop in CAPE in the region (Figure 13). A major rain event around Gan began about 1800 UTC 22 November, which is the subject of tomorrow's summary. |
 |
 |
 |
| Figure 1. Photos looking south and
north at 0448 UTC and NW at 0804 UTC. |
 |
||
 |
 |
 |
| Figure 2. S-PolKa reflectivity PPI for
0216 UTC and reflectivity, velocity, and PID RHI along the
14 degree azimuth at 0206 UTC 22 November. |
 |
 |
 |
| Figure 3. SMART-R reflectivity PPI for
0606-0806 UTC. |
 |
 |
| Figure 4. ARM KAZR reflectivity and
velocity data from 1800 UTC 21 November - 1800 UTC 22
November. |
 |
| Figure 5. Photo looking west at 1145 UTC. |
 |
 |
| Figure 6. Photos looking north at
1041 UTC. |
 |
 |
 |
| Figure 7. Photos looking north at
1107, 1120, and 1150 UTC 22 November. |
 |
| Figure 8. Panorama photo from NW to
NE at 1150 UTC 22 November. |
 |
 |
 |
 |
| Figure 9. Gan soundings for 1800 UTC
21 November and 0000, 0700, and 1200 UTC 22 November. The
0600 UTC sounding aborted below 500 hPa. |
 |
 |
 |
| Figure 10. IMD model analyses at 200 hPa,
700 hPa, and 925 hPa for 0000 UTC 22 November. The red box
indicates the location of the DYNAMO array. |
 |
| Figure 11. Infrared satellite imagery for
0500 UTC 22 November. |
 |
 |
 |
| Figure 12. CIMSS MIMIC Total
Precipitable Water for 0600 20-22 November. |
 |
| Figure 13. Mirai timeseries of CAPE, CIN, and Total
Precipitable Water for October and November. |
| S-PolKa sampled its first major
rain event associated with the onset of the next active
phase of the MJO. Negative velocity potential continues to
spread westward over the Indian Ocean, indicating enhanced
upper level divergence (Figure 1).
The velocity potential now has a distinctive wavenumber-1
structure with the negative portion of the wave centered
over African and the western Indian Ocean, which is
indicative of the MJO entering phase 2 (Figure 2). It took about 30 days
for the MJO to return to this phase. At 0000 UTC 23
November the DYNAMO array was under a large zone of 200
hPa diffluence and 925 hPa convergence (Figure 3). The area over the
northern part of the domain remained very moist, while the
southern array remained very dry (Figure
4). Low-level winds at Gan remained westerly and the upper-level winds easterly (as mentioned in yesterday's report) until 1200 UTC 22 November. However, by 1800 UTC 22 November the winds at all levels began to shift northerly and almost all of the directional wind shear was gone by 0600 UTC 23 November and speed shear was minimal (Figure 5). Additonally, an "onion" sounding occurred at Gan at 0600 UTC 23 November in association with widespread anvil cloud and/or stratiform precipitation. Beginning at 1800 UTC 22 November, a large convective system formed to the west of S-Polka along the eastern edge of a broad convective area to the west (Figure 6). The convective system to the west peaked at approximately 0100 UTC 23 November, after which its stratiform portion appeared in the southwestern portion of the S-PolKa radar domain. Numerous convective cells coming from the NNW quickly collapsed into stratiform precipitation between 1800-2000 UTC 22 November (Figure 7) and resulted in copious amounts of stratiform precipitation that also streamed in from the NNW. The appearance of the stratiform precipitation on KAZR is shown in Figure 8. Convection remained embedded and moved with the stratiform from the NNW until approximately 1800 UTC 23 November (Figure 9). Between 0000-1200 UTC 23 November, the convective cells reached heights of up to 17 km, and brightbands were intense with significant amounts of melting snow and possibly melting graupel (Figure 9). Graupel also appeared at the tops of the high reflectivity peaks of convective cells. By the 1200 UTC 23 November sounding, the winds below 450 hPa regained a westerly component, and the winds above 450 hPa returned to having an easterly component (Figure 5). After 1800 UTC 23 November, a decreased amount of convection was embedded within the stratiform (Figure 10). The event ended in the S-Polka domain within a clearing between three zones of enhanced convection around 2300 UTC (Figure 6) and the layer of westerlies deepening to 350 hPa (Figure 5). Figure 11 compares the echo top and hydrometeor statistics for this rain event and two others: the squall line on 31 October and the convective ring/lines seen on 18 November. In general, both the convective and stratiform echo tops increased in height and frequency from the 31 October event to the 23 November event. The cold-cloud particle types also increased in height over the course of the three events. Although the ice crystals above the melting layer increased in height and frequency through the 23 November event, the rime-ice particle types were at their highest, but least frequent, in the squall line case on 31 October. Rain of all types was most frequent in the 23 November case. Finally, the west and northwest sectors of the radar was contaminated by second-trip echo much of the day after about 1000 UTC 23 November, likely because of strong convection just outside of radar range (Figure 12). Also, the convective/stratiform separation algorithm seemed to strongly favor convection during the period between 0600-1100 UTC (Figure 13). This latter issue seems to be a larger problem at long ranges or in areas not covered by the RHI sectors. |
 |
| Figure 1. 200 hPa velocity potential
anomaly overlaid on the daily infrared satellite
imagery for 22 November. From NCEP
CPC. |
|
| Figure 2. Australian Bureau of Meteorology
MJO phase space from 14 October - 22 November. |
 |
|
 |
| Figure 3. IMD Model analyses at 200 hPa,
700 hPa, and 925 hPa for 0000 UTC 23 November. |
 |
| Figure 4. CIMSS MIMIC total precipitable
water for 0000 UTC 23 November. |
 |
 |
 |
 |
 |
 |
| Figure 5. Gan soundings from 1800 UTC 22
November to 0000 UTC 24 November. |
 |
 |
 |
 |
 |
 |
| Figure 6. S-PolKa reflectivity PPI overlaid
infrared satellite imagery from 1800 UTC 22 November -
0000 UTC 24 November. |
 |
 |
 |
 |
 |
 |
 |
 |
 |
| Figure 7. S-PolKa reflectivity PPI and
reflectivity and radial velocity RHI for 1830-2030 UTC 22
November. |
 |
 |
| Figure 8. KAZR reflectivity and velocity
data from 1800 UTC 22 November - 0000 UTC 24 November. |
 |
 |
 |
 |
 |
 |
 |
 |
 |
| Figure 9. S-PolKa reflectivity PPI and
reflectivity and hydrometeor classification RHI for
0146-0331 UTC 22 November. |
 |
 |
 |
 |
| Figure 10. S-PolKa reflectivity PPI and
convective/stratiform separation for 1800 UTC and 2100 UTC
23 November. |
 |
 |
 |
 |
 |
 |
| Figure 11. S-PolKa echo top (top 2 rows)
and hydrometeor classification (bottom row) statistics for
31 October, 18 November, and 23 November. |
 |
 |
| Figure 12. S-PolKa reflectivity PPI and
infrared satellite imagery for 1130 UTC 23 November. |
|
|
 |
 |
 |
| Figure 13. S-PolKa reflectivity PPI (with
the 8 degree azimuth marked), convective/stratiform
seperation, and reflectivity RHI at 8, 16, and 31 degrees
azimuth for 1100 UTC 23 November. |
| The amplitude of the MJO signal
continues to increase over the western Indian Ocean (Figure 1), with the negative anomaly of
the velocity potential continuing to increase in intensity
(Figure 2). While the bulk of the
negative velocity potential remains over the western
Indian Ocean, a piece of this anomaly continues to spread
eastward along the equator. This is indicative of
increasing upper-level divergence over much of the basin. Coherent 200 hPa anticyclonic gyres have reappeared over the central Indian Ocean, although they are somewhat zonally shifted from each other -- the gyre north of the equator is centered roughly over Sri Lanka (80-85 E) while the southern gyre is centered closer to 75-80 E (Figure 3). The diffluence just north of the DYNAMO array has intensified as the anticyclonic gyres intensify and organize into tighter circulations. At 500 hPa, a strong ridge has developed over India and Bangladesh, with strong southerly-component flow from the Arabian Sea into Pakistan and northwestern India and strong northerly-component flow over western India into the Bay of Bengal. This pattern is likely associated with the winter monsoon circulation. Four cyclonic circulations are now apparent in the flow below 700 hPa--one just north of the Revelle, one in the Arabian Sea, one just west of Diego Garcia, and one off the NE tip of Madagascar. The circulations of the first three cyclonic regions converge directly over the northern part of the DYNAMO array, where low-level westerlies are strong. At 700 hPa a small area of increased relative humidity and decreased geopotential height is directly over the S-Polka site. A large convective region continues to spin up just off the southern coast of India and Sri Lanka, which is considered an "invest" of high probability for development of a tropical cyclone by the Joint Typhoon Warning Center. The NOAA P-3 aircraft flew a convective module just west of the Revelle, and both platforms sampled the convection as it formed into convective lines spiraling into the circulation center (Figure 4). By 0600 UTC, soundings from the Revelle show very little deep-layer shear through 200 hPa except for some variance in the wind direction between 500-700 hPa (Figure 5). Winds from the P3 dropsondes show powerful low-level westerly winds along the Equator (Figure 6), consistent with the 850 hPa map in Figure 3. While the low levels were saturated within the bands of the convective center, the area closer to Diego Garcia was very dry near the surface. Strong shear returned and an "onion" sounding occurred at the Revelle around 1800 UTC 24 November (Figure 5). It was a relatively quiet and breezy morning in the range of S-PolKa as a clearing formed between three convective areas about 0000 UTC 24 November (Figure 7). High clouds and other cloud layers covered the area for most of the day (Figure 8). These numerous cloud layers were seen in the Ka and S bands of S-PolKa (Figure 9) as well as in KAZR (Figure 11). The hydrometeor classification also exibited an interesting stratification of ice particles in these layers, with larger non-melting particles (light blue) lying below smaller ice particles (pink) and over liquid "drizzle" drops (Figure 10). Shear over Gan was minimal up to 400 hPa, but the deep layer shear (e.g. 200-850 hPa) remained intense (Figure 12). Winds at 925 hPa and 200 hPa were both ~15 ms-1 at 0900 UTC 24 November, but in opposing directions (not shown). Large areas of stratiform remained on the fringes and outside of the S-PolKa domain until 0900 UTC. A small line of convection, possibly resulting from the outflow of a storm not visible with the radar, zipped across the northern sectors heading eastward around 0500 UTC (Figure 13). Another line of convection formed NNW of the radar at about 0800 UTC. This line intensified, merged, and was eventually absorbed by the stratiform region that began to stream from the north and was likely from the large line of convective cells to the NE of S-PolKa (Figure 14). Finally, a bowed line of cells merged into a squall line about 1500 UTC 24 November and quickly collapsed into stratiform over the course of an hour (Figure 15). The backward tilt of the reflectivity feature, and the high winds above 8 km, indicate the incredible amount of deep-layer shear this feature experienced. Initially the convection lofted large quantities of ice and dry aggregates to significant levels, but as the convection collapsed significant amounts of melting aggregates and graupel appeared along the brightband. The stratiform region was relatively weak and most of its echo top was at only about 10 km altitude. This squall line was reminiscent of that on 31 October (see summary for that day, especially the last figure in that previous summary). |
 |
| Figure 1. Australian Bureau of Meteorology
MJO phase space diagram for 15 October - 23 November. |
 |
| Figure 2. 200 hPa velocity potential map
for 23 November. Colors are OLR, green contours are
negative velocity potential, and brown contours are
positive velocity potential. |
 |
|
|
|
| Figure 3. IMD model analyses for 200 hPa,
500 hPa, 700 hPa, and 925 hPa at 0000 UTC 24 November. |
 |
 |
| Figure 4. NOAA P3- flight tracks overlaid
the visible satellite imagery and Revelle reflectivity
PPI at 0700 UTC (left), and NOAA P3 flight tracks with
time stamps overlaid on water vapor imagery (right). |
 |
 |
 |
| Figure 5. Soundings from the Revelle for 0000,
0600, and 1800 UTC 24 November. Many of the soundings
between 0600-1800 UTC aborted or were only partial
profiles. |
 |
 |
 |
| Figure 6. NOAA P-3 dropsondes at 0213,
0458, and 0936 UTC 24 November. |
 |
 |
|
|
 |
 |
 |
 |
| Figure 7. Infrared satellite imagery from
0000-1800 UTC 24 November. |
 |
 |
 |
| Figure 8. Photos looking ESE at 0401 UTC,
0611 UTC, and 1213 UTC 24 November. |
 |
 |
 |
 |
| Figure 9. S-PolKa Ka-band (left) and
S-band (right) PPI at 0730 UTC and RHI along 124 degrees
azimuth at 0743 UTC 24 Noember. |
 |
| Figure 10. S-PolKa S-band PID at 124
degrees azimuth for 0743 UTC 24 November. |
 |
| Figure 11. KAZR reflectivity from 0000 UTC
24 November - 0000 UTC 25 November. |
 |
 |
 |
 |
| Figure 12. Gan soundings from 0000-2100
UTC 24 November. There were several aborted or partial
soundings between 1500-1800 UTC. |
 |
 |
 |
| Figure 13. S-PolKa reflectivity PPI for
0446-0646 UTC 24 November. |
 |
 |
 |
 |
| Figure 14. S-PolKa reflectivity PPI and
RHI at 6 degrees azimuth from 0835-0930 UTC 24 November. |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
| Figure 15. S-PolKa reflectivity PPI and
reflectivity, PID, and velocity RHI at 46 degrees azimuth
from 1500-1630 UTC 24 November. |
| The dominant feature in the
Indian Ocean is a large circulation just to the south of
India and Sri Lanka that became increasingly organized
today (Figure 1). During the day
the Joint Typhoon Warning Center acknowledged that this
circulation had a high potential of developing into a
tropical cyclone and labeled it Invest 98B. This
circulation is a stark contrast to the two large bands
of convection that traversed the northern and southern
equatorial portion of the basin on 20
November. Additionally, while the northern band of
convection in the Indian Ocean was highly electrified on
20 November, significant lightning activity was
restricted to a small region along the southern tip of
India today (Figure 2). A few instances
of lightning were also reported near Diego Garcia and
the Revelle.
While the northern 200 hPa anticyclone strengthened and shifted northward over India today, the southern anticyclone has dissipated (Figure 3). 200 hPa diffluence continues to influence the DYNAMO array. At 500 hPa, the winter monsoon circulation observed yesterday has intensified. Confluence at all levels below 500 hPa is occurring over the northern DYNAMO array between the strong westerlies just south of the equator and the burgeoning cyclone circulation. Negative velocity potential continues be centered over the Indian Ocean and slowly propagate along the equator towards the east as a wave number one structure (Figure 4). Additionally, the Wheeler and Hendon Index continues to indicate that the MJO is in phase 2, with enhanced convection in the Indian Ocean (Figure 5). S-PolKa and the Revelle were along the northern edge of a convective band extending into the Invest region throughout the day (Figure 1). Soundings in Gan and at the Revelle continue to indicate moist conditions through 200 hPa (Figure 6). At Gan, deep westerlies up to ~350 hPa and the upper-level speed shear maximized at 0600 UTC with easterly winds increasing from 10 m/s at 300 hPa to 25 m/s at 200 hPa. The westerlies over Gan also had significant speed shear throughout the day as winds increased form 5 m/s at 1000 hPa to approximately 15 m/s at 450 hPa. While the upper level vertical wind shear weakened during the day, the low level speed shear persisted. At the Revelle, the low level westerlies increased in height throughout the day and reached 250 hPa by 1800 UTC. While the westerlies are comparable in magnitude to those witnessed in Gan, the upper level easterlies were 20 m/s weaker than Gan at 0600 UTC. However, by 1800 UTC, the upper level easterlies at the Revelle were 5 m/s faster than those observed over Gan. The Revelle experienced significant convective activity as stratiform and embedded convective lines propagated towards the northeast. One exception to this pattern was an eastward-propagating squall line that passed near the Revelle at approximately 0430 UTC (Figure 7). Time lapse visible imagery overlaid with radar reflectivity from S-PolKa and Revelle indicates that while the convective and stratiform echoes on the radar were moving towards the northeast, cirrus anvils were rapidly spreading towards the southwest (Figure 8). The movement of the cirrus is consistent with observations of strong upper level winds with easterly-components in the Revelle sounding at 0600 UTC (Figure 6). Convection began to move out of the Revelle's domain around 1400 UTC, and only a few anvils were on the fringes of their domain by 2300 UTC (not shown). The day began at S-PolKa with scattered showers that organized into two small east-west lines of eastward propagating convection north and south of the radar by 0600 UTC (Figure 9). The convection was most intense between 0500 and 0800 UTC. During this time, cells in the northern line grew deeper and persisted longer than cells in the southern line. The tallest cell of the day reached a maximum height of 14 km at 0646 UTC, had a small stratiform region, and graupel above the melting level (Figure 10). While most cells were short-lived and trapped below the layer of intense vertical wind shear between 10 and 14 km, this cell was able to penetrate the intense shear layer. The persistent stratiform region that developed beneath the shear layer moved eastward with the convection. Time lapse of radar reflectivity superimposed on visible satellite imagery indicates that as other cells penetrated the upper level shear layer the tops of the storms were rapidly blown towards the southwest. This strong directional and speed shear is consistent with the Gan soundings (Figure 6). While it is likely that strong upper level vertical wind shear prevented most cells from developing deep convection or stratiform regions, one can speculate that the northern line was being more influenced by the outflow from the Invest such that the high cloud in the outflow produced a more favorable environment for the northern line by increasing upper-level moisture or instability. During the afternoon, cells within these two convective lines became more isolated and less robust (Figure 9). By 1000 UTC both lines had dissipated and convection within the S-PolKa domain was extremely isolated. Numerous cloud layers produced overcast skies over S-PolKa throughout the day, which were observed visually (Figure 11) and on the KAZR radar (Figure 12). Between 1200 and 1800 UTC a line of convective cells propagated rapidly towards the east through the southern portion of the S-PolKa domain (Figure 13). This squall line was similar to the convective line observed the previous night (and on 31 October). It was comprised of a number of distinct convective cells that formed a rapidily eastward propagating line of moderate convection, was highly sloped rearward and lacked a robust stratiform region. During this time a rainband began to move through and develop north of the S-PolKa domain. This squall line may have been associated with one of the convective cells moving towards the circulation center (Figure 14). Around 2300 UTC another squall line passed through the S-PolKa domain. However, it was less coherent than the one observed earlier in the evening. Similar to yesterday, the S-PolKa domain experienced significant contamination by second trip echoes as a result of strong convection towards the east (Figure 15). Positive LDR is often associated with second trip echoes and Figure 15 demonstrates one of the most extreme examples of this type of contamination observed today. |
 |
 |
 |
 |
| Figure 1. METEOSAT Infrared satellite
imagery at 0000, 0600, 1200, and 1800 UTC 25 November. |
| Figure 2. WLLN overlaid METEOSAT infrared
imagery at 0901 UTC 25 November. Lightning accumulated
over the previous half and hour. |
 |
 |
 |
 |
| Figure 3. IMD 200, 500, 600, and 850 hPa
synoptic analysis at 0000 UTC 25 November. |
| Figure 4. 200 hPa Velocity Potential
(green contours are negative, brown contours are positive)
and daily IR from the CPC for 24 November. |
| Figure 5. The Wheeler and Hendon MJO RMM
Index for the last 40 days. |
 |
|
 |
 |
 |
 |
 |
 |
| Figure 6. Soundings from Gan and the Revelle for 0000,
0600, 1200, and 1800 UTC 25 November. Many of the
soundings between 0000-1200 UTC were aborted or only
partial profiles. |
 |
 |
 |
| Figure 7. Revelle C-band radar reflectivity at
0401, 0431, and 0501 UTC 25 November. |
 |
 |
 |
 |
| Figure 8. METEOSAT visible imagery
overlaid with S-PolKa and Revelle reflectivity data at 0431, 0501,
0531, and 0602 UTC 25 November. |
|
 |
 |
 |
| Figure 9. S-PolKa S-band reflectivity at
0401, 0601, 0801, and 1001 UTC November 25. |
 |
 |
 |
 |
| Figure 10. S-PolKa S-band PPI of
reflectivity and RHIs of reflectivity, radial velocity,
and hydrometeor classification 0646 UTC 25 November. RHIs
are along the yellow line in the PPIs. |
 |
 |
 |
| Figure 11. Photos looking SSE at
0406 UTC, SE at 0553 UTC, and ENE at 0952 UTC. |
|
 |
| Figure 12. KAZR reflectivity (top) and
velocity (bottom) at Gan from 0000 25 November to 0000 UTC
26 November. |
 |
 |
 |
 |
 |
 |
 |
| Figure 13. S-PolKa S-Band reflectivity
PPIs (top) at 1431, 1446, 1501, and 1516 25 November. RHIs
of reflectivity, radial velocity, and hydrometeor
classification are taken along the yellow line at 1501 UTC
25 November. |
 |
 |
 |
 |
 |
 |
 |
 |
| Figure 14. METEOSAT infrared imagery with
S-PolKa S-Band reflectivity overlaid from 1231 to 1601 UTC
25 November. |
 |
 |
| Figure 15. S-PolKa S-band reflectivity
(left) and linear depolarization ratio (LDR) (right) at
0416 UTC 25 November. |
| The MJO has now advanced into phase
3, with the bulk of the convective signature in the
eastern Indian Ocean (Figure 1).
The amplitude of the signal remains high, and
propagation remains slow although the signal is now
progressing steadily eastward. The central Indian Ocean
shows the most enhanced OLR and strongly negative
velocity potential is now over much of the basin and
continues to slowly build eastward (Figure
2). On 26 November, the large anticyclone at 200 hPa associated with a subtropical ridge shifted eastward over the Bay of Bengal (Figure 3). The region of diffluence previously positioned over the northern DYNAMO array has also shifted eastward, with correspondingly lower wind speeds over Gan. An anticyclonic gyre reappeared in the southern hemisphere, although it is now centered SE of Indonesia so the winds over the Revelle have shifted to being northeasterly at 200 hPa. At 500 hPa, the anticyclone previously over India has shifted eastward, and a band of powerful easterlies between the Equator and 10 N transverse the Bay of Bengal before splitting around the cyclone circulation off the southern tip of India. The meeting of the easterlies with the cyclonic gyres in the eastern Indian Ocean also describes the flow from 700 hPa downward. A small cyclonic gyre NE of Madagascar continued to remain mostly stationary. On 27 November, the 200 hPa northern anticyclonic gyre shifted back towards the west, repositioning the northern array under the main region of diffluence between it and the less organized southern gyre that shifted further eastward towards Indonesia. At 500 hPa and below, the cyclone in the Bay of Bengal moved northwestward, and a new rotation spun up to its southeast between the westerlies, the cyclone circulation, and the easterlies along the Equator. Two lines of convergence formed: one between the cyclone and the gyre on the Equator that created a region of confluence stretching towards the SW from Sri Lanka and one between the westerlies shooting between the gyres and the easterlies that created a line of convergence stretching towards the SE through the Bay of Bengal. The cyclonic gyre NE of Madagascar also shifted eastward, and at 925 hPa the westerlies on its northern side are converging with the easterlies just to the west of the DYNAMO array. The cyclonic disturbance in the Arabian Sea, named Invest 98B, continued to produce the most pronounced high-cloud feature of the Indian Ocean region on 26 November (Figure 4). At 0900 UTC 26 November the Joint Typhoon Warning Center upgraded Invest 98B to Tropical Cyclone 05A, which is moving towards the WNW. A sharp boundary appeared on 26 November between the westerlies, the cyclone circulation, and strong easterlies from the maritime continent, creating a distinct cloud edge in the Bay of Bengal which was highly electrified (Figure 5). Interestingly, the airmass on the non-precipitating side of the boundary is not particularly dry on 26 November (Figure 6). As the cyclone pulled away towards the west on 27 November, the sharp cloud boundary continued to move eastward (Figure 5). A new region of enhanced convection began intensifying at 0000 UTC 27 November south of India and Sir Lanka (Figure 4). By the end of the day, convection was mostly concentrated in the central and eastern Indian Ocean. The feeder bands moving into the large area of convergence to the east and northeast impinged upon the northern portion of the DYNAMO array on 26-27 November, while the southern portion of the array remained quiet (Figure 7). The NOAA P3 aircraft flew another flight near the Revelle on 26 November (Figure 8). The dropsondes revealed dry air near the surface with strong low-level southerlies that turned southwesterly with height at the beginning of the flight (Figure 9). The low levels generally moisten and become fully southwesterly near the Revelle, with wind speeds of 7-10 m/s (15-20 kts). As a result of these large convective regions, high clouds persisted across the S-PolKa area on 26 November (Figure 10). Westerly winds were strong and gusty at S-PolKa. On 26 November winds at the Gan ARM site on 26 November were mostly southwesterly and ranged from ~6-12 m/s (Figure 11). Soundings on 26 November show deep layer westerlies that gained a strong southerly component above 300 hPa (Figure 12), resulting in strong deep-layer directional shear. The upper-level southerlies descended to 400 hPa by 1200 UTC, and by 1800 UTC the winds above 300 hPa regained a strong easterly component, and the low level winds were faster than the upper level winds. On 26 November, a series of isolated convective cells and convective lines zipped across Overnight and into 27 November, the winds picked up dramatically (Figure 11), which combined with heavy rains caused damage all along the atoll (Figure 15). The amount of rain accumulated at the ARM Gan in two hours on 27 November exceeded the total amount accumulated during the rain event on 23 November, which previously held the largest total rain accumulation (Figure 16). The Gan sounding showed a wholesale switch in direction in the winds above 300 hPa between 0000-0600 UTC (Figure 12) as the upper-level winds returned to being easterly by 1200 UTC. The layer of easterlies had decended to 400 hPa by 1800 UTC. Winds between 450-600 hPa had a strong southerly component between 0600-1800 UTC. Additionally, the region between 500-700 hPa was relatively dry from 1200-1800 UTC. Beginning about 2200 UTC 26 November, a series of convective cells appeared and formed into a long line of convection stretching NE to SW by 0000 UTC 27 November (Figure 17) with intense rain and extensive leading and trailing anvil (Figure 18). Timelapse imagery reveals that the propagation of the cells was northeastward, which is the direction the stratiform maintained when the line the cells began to collapse. The stratiform began to move northward at about 0400 UTC . Once the line formed ~0100 UTC, the cells were intense with 40-45 dBZ echo up to ~5 km and 10 dBZ echo up to ~15, penetrating the intense directional shear layer above 12 km. Surface winds behind the line reached ~20 m/s along the radar beam, creating intense convergence in the convective line. The ARM site registered ~17 m/s winds and a drop in temperature of 2 degrees between 0000-0100 UTC (Figure 11). At 0200 UTC, the convergence layer moved upwards to ~2-4 km, and a layer of divergence appeared at the surface just ahead of the line (Figure 17). Intense divergence was also observed above 6 km. Ample amounts of aggregates fell into an intense melting region along the bright band. This melting signature was particularly intense at 0300 UTC, where a > 50 dBZ bright band occurred that the PID identified large amounts of melting graupel and, probably erroneously, a small region of hail. (The algorithm did not know what else to do with these high reflectivities; the algorithm is tuned form midlatitude continental convection, which produce hail and less robust stratiform melting layers.) The height of the dry aggregates also decreased, indicating settling and subsequent melting. Intense convergence continued to occur near 2 km altitude, and the region of surface outflow penetrated farther into the raining region. A sharp peak up to 1013 hPa occurred in the atmospheric pressure just after the collapse of the band near between 0300-0400 UTC 27 November (Figure 11), although the atmospheric pressure for the entire day was higher than 26 November. Two other short lived convective lines passed through the stratiform in the northern sector of the radar, the first at about 0330-0500 UTC and the second at ~0800 UTC. In the afternoon, the sun appeared for the first time in two days, with high cirrocumulus and cirrus holding mostly stationary as the lower clouds rapidly moved eastward (Figure 19). Early evening, the sky darkened to the west, and a v-shaped line of convection began to appear west of the radar about 1100 UTC, moving rapidly towards the east (Figure 20). Just past the radar, the line rapidly intensified and extended into an arc. The velocity field contained bookend vortices and a sharp discontinuity along the leading edge (Figure 21). However, even though the winds were powerful, reaching ~20 m/s at the surface, this line was very squat in the vertical, with echo just reaching 6 km (Figure 22). This convective line did not penetrate into the easterlies above 9 km. but did have a significant backward tilt. Likely because of the low altitude of the convection, the line mostly consisted of warm hydrometeors once it passed the radar. Late in the evening, we received a passing swipe of stratiform precipitation from a large convective region passing to our southeast (Figure 18), which will be the subject of tomorrow's summary. |
| Figure 1. Bureau of Meteorology MJO phase
diagram for 27 November. |
| Figure 2. 200 hPa Velocity Potential (green
contours are negative, brown contours are positive) and
daily IR from the CPC for 26 November. |
 |
 |
 |
 |
|
 |
 |
 |
| Figure 3. IMD 200, 500, 600, and 850 hPa
synoptic analysis at 0000 UTC 26 November and 0000 UTC 27
November. |
 |
 |
 |
 |
|
 |
 |
 |
| Figure 4. Infrared satellite image from
0000 UTC 26 November to 1800 UTC 27 November. |
 |
 |
| Figure 5. Visible satellite image overlaid
with WWLN lightning data at 0600 UTC 26 November and 0600
UTC 27 November. The lightning flashes were accumulated
over 30 minutes. |
|
 |
| Figure 6. CIMSS MIMIC total precipitable
water for 0600 UTC 26 November and 0600 UTC 27 November. |
| Figure 7. S-Polka and Revelle reflectivity
PPI overlaid visible satellite imagery at 1005 UTC 26
November. |
| Figure 8. NOAA P3 flight track overlaid
water vapor imagery at 1100 UTC 26 November. |
 |
 |
 |
| Figure 9. NOAA P3 dropsondes at 0412 UTC,
0556 UTC, and 1009 UTC 26 November. |
| Figure 10. Photo taken towards the SE at
0326 UTC 26 November. |
|
 |
| Figure 11. ARM Gan meteogram for 26-27
November. |
 |
 |
 |
 |
 |
 |
 |
 |
| Figure 12. Gan soundings for 0000 UTC 26
November - 1800 UTC 27 November. |
 |
 |
 |
|
 |
 |
| Figure 13. S-PolKa reflectivity PPI and
convective/stratiform separation for 0745-0945 UTC 26
November. |
 |
||
 |
 |
|
| Figure 14. S-PolKa reflectivity PPI and
reflectivity, velocity, and PID RHI for 1100 UTC 26
November. |
 |
 |
 |
 |
| Figure 15. A downed tree on Gan (~0330 UTC)
and the house it destroyed (taken 0317 UTC 28 November), a
damaged shelter by S-PolKa (0720 UTC), and deep pools of
water on the roads of Hithadoo (0731 UTC) on 27 November. |
 |
 |
| Figure 16. ARM Gan precipitation totals for
23 November and 27 November. |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
| Figure 17. S-PolKa reflectivity PPI and
reflectivity, velocity, and PID RHI at 0000-0300 UTC 27
November. |
 |
 |
| Figure 18. ARM KAZR reflectivity for 26-27
November. |
 |
 |
 |
| Figure 19. Photos facing SW at 1113 UTC,
south at 1205 UTC, and east at 1243 UTC 27 November. |
 |
 |
|
|
| Figure 20. S-Polka reflectivity PPI from
1131-1304 UTC 27 November. The yellow line represents the
36 degree azimuth line. |
| Figure 21. S-PolKa velocity PPI at 1232
UTC. The yellow line represents the 36 degree azimuth
line. |
 |
 |
 |
 |
 |
 |
| Figure 22. S-Polka reflectivity, velocity,
and PID RHI along the 36 degree azimuth for 1239 and 1311
UTC 27 November. |
| The active phase of the MJO signal
continues to propagate eastward in the eastern Indian
Ocean, and is now picking up speed as it also increases in
amplitude (Figure 1). The strongly
negative velocity potential is now centered more in the
eastern part of the basin and the western part of the
Indian Ocean is becoming less negative (Figure
2). Westerly winds remain intense along the equator
at 850 hPa as the northern part of the array remains under
a region of diffluence at 200 hPa (Figure 3). Geopotential height
levels decreased from 500-850 hPa over the DYNAMO array in
conjunction with a gyre on the equator that is
slightly east of the array at 925 hPa and leans westward
with height up to 500 hPa. Cyclone 5A continues to track
NW into the Arabian Sea, and the cyclonic gyre NE of
Madagascar continues to move very slowly westward, with
enhanced westerlies on its northern side. The easterlies
north of the equator continue to converge with the
enhanced westerlies SE of India, and the easterlies south
of the equator converge with northeasterlies that are part
of the circulation over the DYNAMO array near the location
of the Mirai.
Dry air continues to move into the western Indian ocean
from the western Arabian Sea and from east of Madagascar (Figure 4). Although lightning is
occurring along the western, southern, and eastern edges
of the large convective region in the central Indian
Ocean, there is very little lightning within it (Figure 5). A long band of
convection across the Indian Ocean is beginning to
reappear south of the equator, instead of the single giant
cluster of convection seen over the past several days. The area around Gan was generally overcast (Figure 6). Most of our precipitation came from clouds in the western fringes of the major cluster of convection over to our east (Figure 7). Winds were generally westerly up to 400 hPa, with a smaller amount of low-level shear than seen in previous days (Figure 8). The upper-level winds became easterly above 400 hPa by 0600 UTC 28 November and intensified throughout the day. Humidity in the low levels also appeared to be dropping. Most of what passed through the S-PolKa domain was isolated convective cells and a few thin convective lines. However, stratiform was constantly on the southern fringes of the radar and occasionally covered a section of the radar domain. Large amounts of thick mid-level anvil blew in from the south for much of the day (Figure 9). Around 1800 UTC, a line of cells converged into a longer convective line (Figure 10). Winds behind the convective line were intense, although somewhat less intense then yesterday. Turbulence was visible along the leading edge in the ZDR field between 1800-1830 UTC (not shown). The height of the convective cells varied along the line and some were taller than the incredibly short cells witnessed yesterday (Figure 11). A fairly intense descending mid-level jet is also visible behind the line into the stratiform region behind it. However, the stratiform precipitation is notably weak and shallow, reminiscent of 31 October. Meanwhile major mesoscale convection was occurring in other parts of the array. At the Revelle, time lapse imagery of the reflectivity PPI reveals a rotation in the cells overhead from 1400 UTC 27 November - 0000 UTC 28 November as the large convective region moved south of them. This is discussed in more detail in their operational summary. During this time, the winds were easterly down to 700 hPa and the entire profile was moist (Figure 12). As the larger convective cluster moved towards the SW, two MCSs blew up in the center of the DYNAMO array: the first between the Revelle and the Mirai about 2100 UTC 27 November and the second NE of Diego Garcia starting around 1100 UTC 28 November. The Revelle later received large amounts of stratiform precipitation from 0000-1800 UTC 28 November (Figure 7). At the Mirai, winds were easterly through the entire layer, but there was significant unidirectional speed shear in the lower-levels (Figure 13). At 0000 UTC 28 November, winds at the surface were 15 knots and winds at 600 were 30 knots; by 1800 UTC 28 November, the winds were 15 knots at the surface and 50 knots at 650 hPa. The column was moist through most levels. The Mirai was moving north along the 80 E meridian, away from their usual station during that time. At Diego Garcia, the winds were initially mostly easterly at 0000 UTC 28 November, with dry layers at 450-700 hPa and 750-850 hPa (Figure 14). The upper-level moisture descended to 550 hPa by 0900 UTC, and by 1800 UTC the entire profile was moist with low level directional shear between the surface and 800 hPa. The NOAA P-3 flew a boundary layer mission from 0200-1100 UTC, whose dropsondes (from 0217-0344 UTC) show the developing low-level directional shear near the first MCS between Revelle and Mirai (Figure 15). A third MCS developed from the remnants of the second MCS around 2200 UTC 28 November which will be discussed in tomorrow's summary. |
| Figure 1. Australian Bureau of Meteorology
MJO phase diagram for 27 November. |
| Figure 2. 200 hPa Velocity Potential
(green contours are negative, brown contours are positive)
and daily IR from the CPC for 27 November. |
|
 |
 |
 |
 |
| Figure 3. IMD model analyses for 0000 UTC
28 November at 200 hPa, 500 hPa, 700 hPa, 850 hPa, and 900
hPa. |
| Figure 4. CIMSS MIMIC total precipitable
water for 0700 UTC 28 November. |
 |
 |
 |
| Figure 5. WWLN lightining data overlaid
infrared satellite imagery for 0000 UTC, 1400 UTC, and
2100 UTC 28 November. |
 |
 |
 |
| Figure 6. Photos looking east at 0611 UTC,
0936 UTC, and 1205 UTC 28 November |
 |
 |
 |
 |
 |
 |
| Figure 7. S-PolKa and Revelle reflectivity
data overlaid infrared satellite imagery from 1800 UTC 27
November to 0000 UTC 29 November. |
| Figure 8. Gan sounding weekly series from
22-29 November. |
|
 |
| Figure 9. ARM KAZR reflectivity and
velocity from 0000 UTC 28 November - 0000 UTC 29 November. |
 |
 |
 |
 |
 |
 |
 |
 |
| Figure 10. S-PolKa reflectivity and
velocity PPI for 1800-1930 UTC 28 November. |
 |
 |
 |
 |
 |
 |
| Figure 11. S-PolKa reflectivity, velocity,
and PID RHI along the 10 degree azimuth at 1751 UTC and
the 82 degree azimuth at 1927 UTC 28 November, seen in the
first and last panels of Figure 10. |
 |
 |
 |
 |
| Figure 12. Revelle soundings for 2100 UTC 27
November-1800 UTC 28 November. |
 |
 |
 |
| Figure 13.
Mirai soundings from 0000-1800 UTC 28 November. |
 |
 |
 |
| Figure 14. Diego Garcia soundings from
0000-1800 UTC 28 November. |
 |
 |
| Figure 15.
P3 flight track overlaid IR satellite image at 0400
UTC, and the P3 dropsonde for 0345 UTC 28 November. |
| Convection continues to move south
of the equator and stretch further across the entire
Indian Ocean today, with the convection and lightning to
the northwest slowly fading out during the day (Figure 1). With the convective burst
within the DYNAMO array steadily dissipating, most of the
large convection appears to have moved to the eastern side
of the basin. The northern portion of the array is no
longer directly underneath the 200 hPa diffluence region
described yesterday, as the easterlies along the equator
continue to strengthen (Figure 2). The
main zone of diffluence has moved east of the former
location of the Mirai,
which is currently moving off station. The large gyre
associated with a subtropical ridge has returned to be
mostly centered over India, while the southern gyre has
once again lost its coherent structure. The
westerlies at 500 hPa are converging with the easterlies
directly above the Revelle,
in a line of convergence stretching from Cyclone 05A in
the Arabian Sea down to the center of the DYNAMO
array. Geopotential heights remain low over the area
from 500 hPa down, with a
cyclonic rotation positioned directly above the array. At
850 hPa, this cyclonic region has enhanced westerlies on
its northern side and is leaning westward with height less
severely then yesterday. Total precipitable water in the
array has remained about the same as yesterday (Figure 3). Today the DYNAMO array began to quiet down as most of the convection in the region moved to our south and east (Figure 4). At 0000 UTC 29 November a small MCS formed near Diego Garcia, but quickly dissipated by 0600 UTC. Conditions remained moist at Diego Garcia throughout the day, although the shear profile shifted. At 2100 UTC 28 November, Diego Garcia had easterlies down to 750 hPa that turned to westerlies by 950 hPa (Figure 5). By 1800 UTC 29 November, the winds had a strong westerly component up to 700 hPa, a strong northerly component between 700 hPa to 300 hPa, and a strong easterly component above 300 hPa. The speed shear between the low and mid levels increased over the course of the day, and the upper levels and mid-levels dried somewhat. The Mirai experienced scattered convection and stratiform precipitation mainly from the large convective region to its east in the morning as it continued its transit to the north from its previous station (Figure 6). The speed of the westerlies, the amount of deep-layer shear, and the dry air became more dramatic the further north the ship went (Figure 7). The Revelle also experienced scattered convection and stratiform during the day: first, around 0600 UTC, a small squall line; second, around 1500 UTC, a stationary band of convection; and third, near 0000 UTC 30 November, a small amount of stratiform from the convective area to the east of the ship (Figure 8). The shear remained mostly the same, but the low levels dried out over the course of the day, with the dry region elevating from below 700 hPa at 2008 UTC 28 November up to between 550-850 hPa by 2300 UTC 29 November (Figure 9). At S-PolKa today it was overcast with relatively thick upper-level clouds covering the area, as seen in photographs in Figure 10, and in the DOE KAZR data in Figure 11. Within the cloud layer the sounding was moist, but dry air built upward from below through the course of the day along with the westerlies (Figure 12). Low-level unidirectional speed shear also increased during the day between the surface and 600 hPa, and an inversion appeared at 550 hPa by 1800 UTC. Overnight, a small convective line passed through the radar domain (Figure 13), with small convective cells that had relatively high reflectivity in their cores, a divergence signal at their tops, and were just high enough to produce a mixture of ice at their tops (Figure 14). This line did not achieve the level of organization seen over the past few days, although the convective tops were generally higher and wind speeds at low levels were generally lower then what were seen over the past few days as well. Later in the day, layers became apparent in the reflectivity of the S-band of the radar below the layer of ice cloud starting at 5.5 km, similar to the ones described in the 2 October and 6 November reports (Figure 15). These rings were not seen by the Ka band, which strongly indicates that these rings were not created by cloud. The low differential reflectivity indicates that these layres are produced by Bragg scattering, likely through the gradients in humidity seen in the sounding described by Jennifer Davison and others at the University of Illinois (see the 2 October and 6 November report for further examples). |
 |
 |
 |
 |
| Figure 1. Infrared satellite imagery
overlaid with WWLN lightning data from 0000-1800 UTC 29
November. |
 |
 |
 |
|
| Figure 2. IMD model analyses for 0000 UTC
29 November at 200 hPa, 500 hPa, 700 hPa, and 850 hPa. |
| Figure 3. CIMSS MIMIC total precipitable
water for 0700 UTC 29 November. |
 |
 |
 |
 |
| Figure 4. Infrared imagery for 0000-1800
UTC 29 November. |
 |
 |
 |
| Figure 5. Diego Garcia soundings for
1200-1800 UTC 29 November. *Note that at the time of this
report, the 0000-9000 UTC soundings were unavailable. |
| Figure 6. Mirai reflectivity PPI for 29 November. |
 |
 |
 |
 |
| Figure 7. Mirai soundings for 0000 UTC 29 November
- 0000 UTC 30 November. |
 |
 |
|
| Figure 8. Revelle reflectivity PPI for 0629 UTC,
1459 UTC, and 2329 UTC 29 November. |
 |
 |
 |
| Figure 9. Revelle soundings for 2000 UTC 28
November, 1400 UTC 29 November, and 2316 29 November.
*Note that at the time of the writing of this report, the
2300-01200 UTC soundings were unavailable. |
 |
 |
|
| Figure 10. Photos looking ESE at 0421 UTC
and 0812 UTC 28 November. |
| Figure 11. ARM KAZR reflectivity for
29November. |
 |
 |
 |
 |
| Figure 12. Gan soundings from 0000-1800
UTC 29 November. |
 |
 |
 |
 |
| Figure 13. S-PolKa reflectivity PPI for
2200 28 November -0100 UTC 29 November. |
 |
 |
 |
| Figure 14. S-PolKa reflectivity, velocity,
and PID RHI for 0130 UTC, along the 110 degree azimuth
seen in Figure 13. |
 |
 |
|
 |
| Figure 15. S-PolKa S-band reflectivity PPI
(upper); reflectivity RHI for the Ka and S-band radars
(middle); and differential reflectivity (ZDR) and
polarimetric correlation coefficient (rhoHV) RHI for the
S-band (lower) at 0630 UTC 29 November, along the 10
degree azimuth. |
| The region has quieted down
dramatically as dry air moves into the northern DYNAMO
array (Figure 1). Winds with strong
easterly component at 200 hPa continue to increase between
10 N and 10 S, especially the east-southeasterlies just
north of Gan (Figure 2). The 200
hPa subtropical ridge north of India has flattened
somewhat and shifted more to the west, and the
anticyclonic gyre to the south of the equator has reformed
and shifted more to the west as well. A 500 hPa
circulation is still apparent directly over the Revelle at the
convergence of easterlies and westerlies along the
equator, also another cyclonic circulation is located to
its southeast. The two circulations combine to create
strong 500 hPa northwesterlies over the DYNAMO array. The
winter monsoon circulation is still apparent over India at
this level, but its penetration into the tropics is
hindered by the strong easterlies that extend up to 15 N
on its eastern branch. At 850 hPa, the westerly winds are
intense just south of the equator. A cyclonic circulation
south of Diego Garcia has deepened somewhat since
yesterday, and another has formed SE of Sri Lanka between
the intense westerly winds and strong easterlies to its
north. Lightning is once again active in the line of
convection east of Sri Lanka, the circulation south of
Diego Garcia, and in the remnants of Cyclone 05A that are
still spinning down in the Arabian Sea (Figure 3). These changes in the
synoptic-scale are consistent with the MJO active phase
moving farther into phase 3 (Figure 4;
please note that this image is 2 days old). The basin is
still showing highly negative velocity potential,
indicating that upper-level divergence is still relatively
high across the region (Figure 5;
please note that this image is 1 day old). As dry air moved in, convection quickly cleared out of the DYNAMO array (Figure 6). In the northern portion of the array, Gan had westerlies and dry air up to 400 hPa for most of the day, with moister air above that level (Figure 7). The Revelle had westerlies and dry air only up to 600 hPa, and was very moist above that level. Diego Garcia had westerlies up to 800 hPa, northwesterlies above that to ~300 hPa, and easterlies above that level (Figure 8). The soundings began moist at 0000 UTC, but dry air appeared to build downward from above beginning at 0600 UTC, eventually leading to very dry layers between 400-600 hPa and above 350 hPa at 1800 UTC. By this time, the winds in the dry layer gained a stronger westerly component as well. The P3 aircraft flew into a short-lived MCS south of Diego Garcia (Figure 9). The dropsondes show the drastic differences between the environment within the convective region and the area just to its north (Figure 10). The air within the convection was fully saturated, had deep easterlies, and was unstable. The region just to the north was much drier more stable, and had deep northwesterlies. Back at S-PolKa, the radar was mostly clear (Figure 11), except for a few distinct lines of cumulus to our NE that occasionally produced short-lived precipitating convective cells (Figure 12 and Figure 13). Note that the precipitating cells were arising out of lines of nonprecipitating cloud seen via Bragg scattering on S-PolKa. These cloud lines were oriented NW to SE throughout the day and propagated towards the SE. Although the deepest of the precipitating cells were not particularly tall (10-12 km), the convective cores had high reflectivity, healthy updrafts, and large divergence signals aloft. These cells produced a fair amount of aggregates and a small amount of graupel when they were deep enough to go above the melting level. Time-lapse visible satellite imagery indicates that the high thin clouds east of the radar were moving towards the NW, and the occasional low cumulus clouds were quickly moving towards the SE (Figure 14). A few small clouds that seemed to be a product of wave propagation were briefly overhead, especially during the early morning hours (Figure 15). Towards the end of the day, some thicker high clouds reappeared over the area from the large convective area to our east (Figure 16). |
| Figure 1. CIMSS MIMIC total precipitable
water for 0700 UTC 30 November. |
 |
 |
 |
| Figure 2. IMD model analyses for 0000 UTC
30 November at 200 hPa, 500 hPa, and 850 hPa. |
| Figure 3. Lightning overlaid infrared
satellite imagery for 0600 UTC 30 November. |
| Figure 4. Australian Bureau of Meteorology
MJO phase diagram for 28 November. |
| Figure 5. 200 hPa velocity potential
anomaly overlaid the infrared imagery from the CPC. |
 |
|
|
|
 |
 |
| Figure 6. Infrared satellite imagery from
1500 UTC 29 November - 21 UTC 30 November. |
|
 |
| Figure 7. Gan and Revellesoundings at 1200 UTC 30
November. |
 |
 |
 |
 |
| Figure 8. Diego Garcia soundings for 30
November. |
| Figure 9. NOAA P3 flight tracks overlaid
visible satellite imagery at 0700 UTC 30 November. |
 |
 |
| Figure 10. NOAA P3 dropsondes for 0421 UTC
and 1055 UTC 30 November. |
 |
 |
 |
| Figure 11. Photos looking ESE at 0347 UTC,
ENE at 0557 UTC, and SE at 1103 UTC 30 November. |
 |
||
 |
 |
 |
| Figure 12. S-PolKa reflectivity PPI (top)
2146 UTC, and reflectivity, velocity, and PID RHI at 34
degrees azimuth (bottom) at 2153 UTC 29 November. |
 |
||
 |
 |
 |
| Figure 13. S-PolKa reflectivity PPI (top)
0646 UTC, and reflectivity, velocity, and PID RHI at 34
degrees azimuth (bottom) at 0653 UTC 29 November. |
 |
| Figure 14. Visible satellite imagery for
0600 UTC 30 November. |
| Figure 15. Photo taken towards the ENE at
0557 UTC. This image is a zoomed-in version of the middle
photo of Figure 11. |
| Figure 16. ARM KAZR reflectivity for 30
November. |