Bartlett C. Hagemeyer
National Weather Service
Melbourne, Florida
- Early experience with WSR-88D radars suggests that as this enhanced observing technology
is located in new areas, meteorologists may expect to see many "new" phenomena.
The Melbourne, Florida WSR-88D has been operating since October 4th, 1991, and was the
first operational Doppler radar located in a maritime subtropical environment in the
United States. Forecasters anticipated there would be some surprises in store, and that
perhaps some of the conceptual models associated with radar echoes observed on Doppler
radars in other parts of the country might not be directly applicable to our area. We had
to wait only a month for our first surprise; a supercell over the Atlantic Ocean on 6
November 1991. This note will serve as a brief introduction to this storm. A more in-depth
investigation is planned, including data from similar supercell storms that have been
collected since this first storm was observed.
2. Synoptic Overview
- The 0.5ø Base Reflectivity product for 1510 UTC 6 November 1991 from the Melbourne
(MLB) WSR-88D (Fig. 1) shows
the storm in question near the time of its peak intensity about 65 miles east of Cape
Canaveral (arrow at "A"). The maximum reflectivity displayed in Fig. 1, near the
southeastern flank of the storm, is 60 dBZ. Reports from weather buoys 41009 and 41010
east of Cape Canaveral, and two ships (KCLZ and NCAW) in the vicinity are also plotted on
Fig. 1.
- A composite synoptic analysis for 1200 UTC 6 November 1991 is shown in Fig. 2. A surface trough
extended from near Vero Beach, northeastward over the Atlantic Ocean (the trough axis is
also drawn on Fig. 1). Offshore reports suggest the showers and thunderstorms seen in Fig.
1 were located primarily along, and to the south of, this surface trough. A stationary
front was located south of Florida, and strong low-level thermal and moisture gradients
were established from northwest to southeast across the peninsula. Temperatures and
dewpoints were below freezing over northwest Florida, while dewpoints off the southeast
Florida coast were near 70ø F. Water temperatures measured at the buoys were 26 C (79 F).
- A 500 mb trough extending from the Ohio Valley to the central Gulf Of Mexico was
approaching the Florida Peninsula from the west. Relatively lower pressure was offshore
from West Palm Beach at 850 and 700 mb, in an area of widespread convection. This was
indicative of incipient low pressure development in that area. An 850 mb warm, moist
tongue extended from the Florida Straights to offshore of Cape Canaveral. There were no
jets at or below 500 mb. The most outstanding feature of the dynamic environment was the
sub-tropical jet lying across the peninsula and above the surface trough with the
right-rear quadrant of an 86 kt speed max approaching the area of the storm. 1501 UTC IR Satellite
- A sounding and hodograph representative of the thunderstorm environment 65 nm east of
Cape Canaveral (XMR) at 1510 UTC were produced by adjusting the 1100 UTC XMR sounding data
with the Hart and Korotky (1991) SkewT/Hodograph Analysis and Research Program (SHARP).
The results are shown in Fig.
3a-b and Fig. 3c-d which
compare the XMR sounding and modified Gulf Stream sounding. Actual measurements of surface
temperature and winds from Buoy 41010 south of the trough, and representative of storm
inflow, as well as temperature and dewpoint estimates from detailed upper air analyses,
were used to modify the Cape Canaveral sounding to produce a sounding and hodograph more
representative of the later environment over the Gulf Stream.
- The modified sounding revealed an environment conducive to supercell thunderstorm
development with moderate (almost high) Convective Available Potential Energy (CAPE) and
strong shear (Bluestein 1993). CAPE (B+) was estimated at 2424 J/kg, 0-6 km shear was 25
m/s, and 0-3 km storm-relative (SR) helicity was 226 m2/s2 (based on an observed storm
motion of 10 kt from 272ø). By comparison, the unmodified Cape Canaveral sounding was not
conducive to supercell development (CAPE: 222 J/kg, 0-6 km shear less than 25 m/s, and SR
helicity, based on the same storm motion: 102 m2/s2).
- In summary, the storm developed along a strong low-level convergent boundary with
copious low-level moisture and CAPE, strong low-level shear and helicity, and under the
right-rear quadrant of a strong upper-level jet max.
3. WSR-88D Observations
- The storm was identified and tracked by the WSR-88D from 1412 to 1557 UTC. It exhibited
varying degrees of rotation from 1423 to 1545 UTC; peak rotation was observed at 1510 UTC.
The storm was quite electrically active, registering 167 strokes on the Lightning Position
and Tracking System (LPATS) during the hour between 1430 and 1530 UTC (Fig. 4). The storm slowed and
moved to the right of the mean wind during its life cycle. This was evident in the
observed lightning track and in the WSR-88D storm tracking algorithm past positions of the
storm. Output from the algorithm showed initial movement of 19 kt from 240ø at 1418 UTC,
which eventually became 11 kt from 302ø by 1557 UTC. A detailed analysis of storm
evolution and structure is underway, but this brief note will concentrate on the storm at
the time of its peak intensity, around 1504 to 1510 UTC.
- An 8:1 magnified four-panel display of 0.5ø Base Reflectivity, Base Velocity, One Hour
Precipitation, and Echo Tops products at 1510 UTC, centered on the area of the storm, is
shown as Fig. 5a-b and Fig.5c-d. Note the distinct
pendant echo with strong reflectivity gradient on the inflow flank of the storm in Fig. 5a-b. The previous
reflectivity product at a.)1504 UTC
and b.)1504 UTC revealed a near
classic "hook echo." A mesocyclone (the red/green couplet) is clearly evident in
the corresponding base radial velocity product (Fig. 5a-b - right side), centered
over the weak echo inflow notch. The rotational velocity was approximately 31 kt at 70 nm
indicating a mesocyclone of moderate intensity. The hook echo was more developed at 1504,
but the mesocyclone was stronger and the peak rotational velocity for the life of the
storm was observed at a.)1510 UTC
and b.)1510 UTC. Correcsponding
reflectivity panels a.)1510 UTC
and b.)1510 UTC.
- The One Hour Precipitation product Fig.5c-d.(lower left) indicates a
maximum rate of 3.7 in/hr. This seems reasonable given copious low-level moisture and
slowing storm speed, and it is consistent with some of the very heavy rainfalls which
occur along the Florida East Coast in the fall with flow off of the Atlantic. Those rains
may be associated with similar storms.
- The Echo Top product Fig.5c-d
(lower-right) indicates a storm top of 35,000 to 40,000 ft. In this case the maximum storm
top was displaced downwind of the updraft core due to strong shear between 30 and 35,000
ft. Indeed, the reflectivity cross-section through the storm at 1516 UTC (Fig. 6a) shows a storm top of
just under 30,000 ft. A distinct Weak Echo Region (WER) can clearly be seen on the
southern flank of the storm. Maximum reflectivity of 58 dBZ was observed at 16,000 ft,
directly above the WER. The highest reflectivity value aloft peaked at 63 DBZ at 1504 UTC.
- The corresponding base velocity cross section for 1516 UTC (Fig. 6b) shows the deep
circulation within the storm. Data begin at 7,000 ft due to the beam height at 70 mi.
Maximum inbound and outbound velocities were 22 and 39 kt, respectively. The circulation
center corresponds with the WER in Fig. 6a, and was undoubtedly stronger below 7,000 ft.
The cross sections illustrate the unusual nature of this fully developed, but
"short", thunderstorm.
- While the sounding indicates an Equilibrium Level (EL) of 42,500 ft, the storm topped
out around 35,000 ft, near the level of the 86 kt speed max at 250 mb, and downwind of the
updraft core.
4. Conclusion
- There is not universal agreement on supercell criteria, and a variety of supercell types
have been described (Doswell, 1985). Although the height of this storm may have been
considerably lower than that usually associated with Midwestern supercells, the Gulf
Stream storm developed in an environment with buoyancy and shear conducive to supercell
development. It clearly classifies as a supercell based on many criteria, including:
longevity (over 90 min), the existence of a WER, a low-level pendant or hook echo on the
right flank of the storm with strong inflow gradient, persistent cyclonic circulation with
an occasional mesocyclone centered on the WER, movement to the right of the mean wind, and
excessive rainfall. While the extent of lightning activity has not been described as a
supercell characteristic, this storm was very active.
- The storm described here is interesting from a purely scientific standpoint, especially
for those specializing in thunderstorm morphology and dynamics, as it resembles a
"baby" midwestern high-precipitation (HP) supercell. However, such storms also
pose hazards to marine and coastal interests and may be indicative of a type of
low-topped, but dangerous, tornado producing storm that occurs in the southeastern states,
primarily from Fall through Spring.
- Forecasters at Melbourne have subsequently observed many other cases of significant
rotation in supercell and non-supercell thunderstorms over the warm waters of the Gulf
Stream, nearer shore, and in different flow patterns. The 6 November 1991 storm is
representative, and a more detailed investigation is underway. As more WSR-88Ds are
deployed in different environments forecasters may look forward to many other interesting
observations; and they should be prepared to adjust established storm models to fit their
local environment.
5. References
Bluestein, H. B., 1993: Synoptic-Dynamic Meteorology in Midlatitudes. Vol. II:
- Observation and Theory of Weather Systems. Oxford University Press,
- New York. 594pp.
Doswell, C. A. III, 1985: The operational meteorology of convective weather
- Volume II: storm scale analysis. NOAA Tech. Memo. ERL ESG-15.
Hart, J. A., and J. Korotky, 1991: The SHARP Workstation V1.50. National
- Weather Service, Eastern Region HQs. Bohemia, NY.
Figure 1. Base reflectivity (0.5ø) product for 1510 UTC 6 November 1991, from
Melbourne WSR- 88D with annotations of surface trough, data buoy weather reports (41009
and 41010) and ship weather reports (NCAW and KCLZ). Air temperature is plotted to the
upper-left of station model, water temperature to the lower-left.
Figure 2. Composite synoptic analyses for 1200 UTC 6 November 1991. Surface freezing
line shown as dotted line, 850 mb warm/moist tongue as shaded area, isodrosotherms as thin
solid lines, 200 mb jet as bold line with arrowhead and jet max as bold "X."
Fronts and troughs shown by conventional notation.
Figure 3a-b. Skew-T, Log-P (3a) and hodograph (3b) representative of Gulf Stream storm
environment at 1500 UTC 6 November 1991.
Figure 4. Plot of total lightning strokes (167) recorded during the one-hour
electrically active phase (1430-1530 UTC 6 November 1991) of the Gulf Stream storm.
Figure 5. Four-panel 8:1 zoom of 0.5ø Base Reflectivity (upper- left), 0.5ø Base
Velocity (upper- right), One-Hour Precipitation (lower-left), and Echo Top (lower-right)
products centered on the Gulf Stream storm.
Figure 6a-b. Reflectivity (6a) and velocity (6b) cross-sections through the Gulf Stream
storm. Section is roughly perpendicular to the radar beam and through the WER for 1516 UTC
6 November 1991.
RADAR OBSERVATIONS OF OTHER GULF STREAM SUPERCELLS
12/27/91 04:10 UTC Composite
Reflectivity, Base Reflectivity
and Velocity, 1-Hour Rainfall
and Echo Tops.
2/20/92 01:06 UTC Composite
Reflectivity and Reflectivity
Cross Section.
3/7/92 23:08 UTC Composite
Reflectivity, Storm Relative
Velocity (.5 degrees), and Reflectivity
Cross Section.
4/6/95 12:57 UTC Composite
Reflectivity and Storm Relative Velocity Mapand Composite Reflectivity and Storm
Relative Velocity Closeup.
