Convection
initiation(top)
Generation of atmospheric bores by colliding density currents
David Kingsmill (Desert Research Institute) and Andrew
Crook completed a study of atmospheric
bores observed during the Convection and Precipitation
Electrification Experiment (CaPE)
in central Florida. These bores were
generated by colliding density currents and are often important
in generating new convection or reinvigorating existing convection.
The study examined whether the strength and structure of
the bore could be predicted given knowledge of the colliding
density currents. It was found that the collision of density
currents of similar strength generated two bores with wave-like
characteristics propagating away from the collision point.
Collisions of unequal density currents tended to generate
one bore with mass transport characteristics propagating
in the direction of the stronger density current.
Land surface variability and dryline convection
Stanley Trier, Kevin
Manning and Fei Chen (RAP) collaborated
on a multiscale study of convective initiation in a mesoscale
model, in particular the sensitivity to soil moisture preconditioning
by antecedent precipitation. The soil moisture fields in
the model were initialized by a long-term (several months)
high-resolution land data assimilation system (HRLDAS) designed
by Chen and Manning. Using
this detailed land-surface information,
Trier, Manning, and Chen
analyzed and simulated a convective outbreak along the
dryline over the Southern Plains that
occurred on 19 June 1998. The MM5 model, initialized with
40-km RUC-II analyses and nested to a grid
spacing as fine as one km in the horizontal, accurately replicated
the timing and location of convection initiation within the
updrafts of flow-parallel rolls in the planetary boundary
layer (PBL) during the late afternoon.
|
|
| Figure
3: PBL rolls
localized near the dryline and simulated by the MM5
model, initialized with a 40-km RUC-II analyses and
nested as fine as 1-km grid spacing in the horizontal.
The model accurately replicated the timing and location
of the convection initiation within the updrafts of
PBL
rolls
in the planetary boundary layer. |
These PBL rolls were localized near the dryline to a region
where the PBL top was elevated due to persistent lifting
within a solenoidal circulation induced by differential sensible
heating. Sensitivity studies indicated that accurate specification
of soil moisture gradients in the initial condition was critically
important to the timing and location of subsequent convection
initiation through its effect on the partitioning of surface
sensible and latent heat fluxes, which influenced the mesoscale
circulations along portions of the dryline. These findings
have important implications for operational NWP, since a
simulation that used the coarse ETA model soil moisture field
was not able to forecast accurately timing and location of
convection initiation.
Use of RUC-II analyses and forecasts guides nowcasts of
deep convection
Trier and David
Ahijevych continued an ongoing collaboration
with Cindy Mueller, Dan Megenhardt, and Nancy Rehak (all
of RAP) using RUC-II analyses and forecast products to assess
the potential for growth/dissipation of mesoscale regions
of deep convection over zero to three hour periods. Efforts
used fuzzy logic to combine output from an algorithm that
diagnoses the depth of thermodynamic instability (developed
the previous year) with RUC-II 3-h convective precipitation
forecasts and a large-scale forcing field derived from RUC
model output. Combining these fields was motivated by an
examination of four test cases that occurred during the 2002
warm season. The cases selected comprised different meteorological
situations in which mesoscale regions of thermodynamic instability
were diagnosed, but the areal coverage and
organization of deep convection, and thus disruption to commercial
aviation, were highly variable. Combining an interest field
based on large-scale forcing with the depth of thermodynamic
instability and RUC-II
3-h forecasted convective precipitation interest fields preserved
guidance for forecast mesoscale regions of organized convection.
|
|
| Figure 4: RUC-II 3-h
convective precipitation forecasts and a large-scale
forcing field derived from RUC
model output. |
At the same time, it reduced areal coverage
of false alarms most prevalent in areas of thermodynamic
instability that
were removed from fronts or other linear regions of large-scale
forcing (see Fig. 4). The combined interest field was implemented
and later evaluated during the FAA Regional Convective Weather
Forecast (RCWF) 2003 summer exercise. Statistical calculations
for this period confirmed that the combined interest field
provided better guidance for short-term convective forecasts
than did individual interest fields alone.
Convective
evolution in complex mesoscale environments(top)
The Bow Echo and MCV Experiment (BAMEX)
The field phase of BAMEX, directed by Christopher
Davis and Morris
Weisman, took place from 20 May to 6 July
2003 and was based at MidAmerica Airport near St. Louis.
MMM scientists Stanley Trier,
David Ahijevych, George
Bryan (ASP), David Dowell, and Jason
Knievel participated.
Trier assisted in organizing
the experiment and served as both a forecaster and
nowcaster during the field phase. Ahijevych,
Bryan, Knievel, and Diana Bartels (NOAA Forecast
Systems Laboratory) coordinated the locations of dropsondes. Dowell coordinated
the deployment of the mobile ground based observing
system. BAMEX is a collaboration among principal investigators
at NCAR, National Severe Storms Laboratory (NSSL),
National Weather Service (NWS), and roughly a dozen
universities. The goals of this experiment are (1)
to obtain kinematic and thermodynamic documentation
of the development of system-scale circulation features
behind the leading convective line in maturing and
decaying mesoscale convective systems (MCSs); (2) to
understand mechanisms of convective regeneration near
mesoscale convective vortices (MCVs) and the dynamics
of MCV intensification that appear critical for multi-day
events; (3) to understand the cause of damaging surface
winds in bow echoes; and (4) to assess predictability
of long-lived MCSs and their effects on weather. Using
two Doppler-radar equipped P-3s, a dropsonde aircraft,
and a movable ground-based observing system consisting
of a wind profiler, an acoustic sounder radiometer,
a mesonet, and soundings, data were collected during
eighteen intensive observing periods. Twenty systems
were sampled: eight severe bow echoes, seven non-bowed
systems, and five mature MCVs (those that survived
at least six hours beyond the decay of the convective
system within which they formed). Initial findings
include a wealth of vortices on many scales, observations
of upscale growth of vortices, and the observation
of deep rear-inflow layers in severe MCSs.
|
|
Figure
5: Example of "quad-Doppler" retrieval
in a bow-echo case (June 10, 2003). Both P-3s
flew in tandem
on opposite sides of the convective line. Each
is capable of dual-Doppler retrieval as each
radar scans rapidly fore, then aft, creating
a series
of points in space where the two beams intersect.
|
Figure
6: Composite map showing locations of
nearly all dropsondes collected in BAMEX (about
440), color-coded by
intensive observing period (IOP).
|
Following the field phase of BAMEX, Trier collaborated
with Davis in a study
of MCVs and their association with redevelopment of
deep convection. Of the five
mature MCVs, three were associated with the retriggering
of deep convection during subsequent diurnal cycles.
The kinematic and thermodynamic structure of these
five long-lived cases is being examined using a combination
of dropsonde, profiler, and mobile sounding data. These
data allow unprecedented documentation of MCV circulations
and suggest the importance of the interaction of the
MCV circulation with the environmental vertical shear
in focusing subsequent deep convection.
|
|
| Figure
7: An example of the lower tropospheric
circulation within one of the more intense MCVs that
occurred in BAMEX. Warm advection in the right downshear
quadrant of the MCV (left panel) helps initiate new
convection near the Arkansas/Missouri/Tennessee border
region (right panels), while strong subsidence inhibits
deep convection downshear of the MCV center in northwestern
Arkansas. |
Figure 7, above, shows an example of the lower tropospheric
circulation within one of the more intense MCVs that
occurred in BAMEX. Warm advection in the right downshear
quadrant of the MCV (left panel) helps initiate new
convection near the Arkansas/Missouri/Tennessee border
region (right panels), while strong subsidence inhibits
deep convection downshear of the MCV center in northwestern
Arkansas.
Severe Thunderstorm Electrification and Precipitation
Study (STEPS)
Charles
Knight, with William
Hall and Jay Miller,
analyzed data collected during the Severe Thunderstorm
Electrification
and Precipitation Study (STEPS), which took place
near Goodland, Kansas from 19 May to 16 July 2000.
This
study was conducted to help improve knowledge of
the interactions between kinematics, precipitation
production,
and electrification in thunderstorms on the High
Plains, which in turn should contribute in significant
ways
to improved forecasts. Knight,
Miller, and Hall completed
a study of convection within anvil precipitation
of a small mesoscale convective system observed during
STEPS. This convection evidently is generated by
instability
created beneath the melting layer, but it extends
through the anvil to 12 km. Its depth may result
from instability
created by the saturation of the air by the falling
precipitation, and this effect may be an important
factor for precipitation production in some mesoscale
convective systems.
Sarah Tessendorf, Kyle Weins, and Steven Rutledge
(all from Colorado State University), along with
Miller,
neared completion of their study of a supercell observed
29 June 2000 during the STEPS field campaign near
Goodland, Kansas. They examined radar and lightning
data to document
the evolution of storm kinematics, microphysics,
and electrical properties over a nearly four-hour
period.
During the storm's early phase there was little lightning
activity or evidence of hail. After intensifying
and making a 35-degree right
turn,
both storm hail volume and lightning activity rapidly
increased. Significant hail production and peaks
in lightning activity were well correlated with increases
in updraft speed.
|
|
| Figure
8: Swaths
of fallout locations (small black +'s) for hail
diameters exceeding (a) 3 cm with radar reflectivity
at 3 km, and (b) 2 cm with maximum updraft in each
vertical column. Reflectivity and updraft follow
the color scales on the right. |
Calculated growth
trajectories from small embryonic particles placed
in the Doppler wind fields indicated that the upwind
side of the updraft was an important source region
for embryos that grew to hail. Substantial increases
in the volumes of updraft and cyclonic vorticity were
required before hail could be grown in the trajectory
model.
Orographic
effects(top)
Orographic control of rainfall coherence
David Ahijevych studied
warm-season rainfall with spatially-averaged
diurnal composites of
precipitation frequency. At first glance, rainfall
patterns during 2002 looked very different from those
of the previous six years. Eastward propagating rainfall
systems were confined poleward of about 42°N
and did not evince similar coherent nocturnal propagation
characteristics as in other years. However, upon recomputing
the diurnally averaged rainfall frequency in a coordinate
system shifted to place the continental divide at a
constant longitude poleward of 40°N, Ahijevych found
that the characteristic timing location of precipitation
reappeared. This solidified the idea that rainfall
propagation is phase locked to the orography
and diurnal cycle.
Convection initiation over the Rocky Mountains and
generation of mesoscale convective systems
Andrew Crook and Donna Tucker (University of Kansas)
continued a study on convection initiation over heated
topography. A linear model of the flow induced by both
terrain (idealized) and elevated heating was developed
and compared with the results from nonlinear simulations.
|
|
| Figure
9: Swaths of fallout locations (small
black +'s) for hail diameters exceeding (a) 3
cm with radar reflectivity at 3 km, and (b) 2
cm with maximum updraft in each vertical column.
Reflectivity and updraft follow the color scales
on the right. |
Figure 9 shows the vertical velocity that develops
at a Froude number equal to 1 and a non-dimensional
heating rate of 0.5 (upper panel, linear solution,
lower panel, numerical results). The two solutions
agree well, except in the lee of the obstacle where
the heating reduces the stability in the numerical
simulations leading to enhanced vertical velocity.
Simulations have also been performed for flow past
a heated ridge. It is shown that the lifting is enhanced,
leading to preferential development of convection,
when the flow is along a heated ridge rather than across
the ridge.
Long
time scale dynamics of mesoscale convective systems (top)
Warm weason rainfall episodes
Richard Carbone,
with John
Tuttle, David Ahijevych, and
Christopher Davis,
examined the systematic variability of warm season
rainfall
over the continental U.S. at timescales
from semi-diurnal to inter-annual while retaining a
spatial resolution of approximately four km. The period
of record and associated statistics have been extended
to seven years (1996-2002), and are summarized briefly
in Movie 1, showing the mean diurnal cycle for June.
|
|
| Movie 1: The
systematic variability of warm season rainfall
over the continental
U.S., at timescales from semi-diurnal to inter-annual
for 1996-2002,
is
summarized
briefly
in the movie. |
Mouse over
image to begin movie. Alternately,
you may download
the animation. |
A detailed
examination of the full seven-year period showed that
semi-diurnal forcing at continental
scales is a myth. Semi-diurnal signals are indeed significant
in the harmonic analyses of summertime precipitation.
However, they are essentially unrelated to atmospheric
tidal scales of motion. The first source of semi-diurnal
forcing is a delayed-phase signal from non-local
diurnal forcing
and the subsequent propagation of rainfall
systems for 12 or more hours. The second source is
a semi-diurnal signal from localized sea/lake/land
breezes, principally along the Gulf of
Mexico and Atlantic Ocean coasts of the southeastern
U.S. Unlike the first example, breezes are a genuine
semi-diurnal forcing, and
are inherently local.
In midsummer (June, July, and
August), there exists a well-defined corridor of
precipitation episodes, which is most prominent across
the central
U.S. Such corridors are vaguely analogous to cool
season storm tracks and are, at best, only weakly
related to fronts and cyclones. In any given year,
a corridor
will experience excessive cumulative rainfall while
nearby regions are often well below normal. Persistence
in the corridor location varies from one week to
more
than one month with significant variability on
the intra-seasonal timescale.
|
|
| Figure
10: Well-defined warm season predipitation corridors
are most prominent across the central U.S. Persistence
in the corridor location varies from one week to
more
than one month with significant variability on
the intra-seasonal timescale. |
This suggests predictability,
in the probabilistic sense, for ranges up to intra-seasonal.
Year-to-year variability of the corridor position appears
to be systematic and may be related to the phase of
ENSO. Figure 10 shows the July 1998 corridor as well
as its position in 1997 and 1999.
Ahijevych continued to document warm season precipitation
patterns across the U.S. by applying simple transformations
to national radar reflectivity composites. These analyses
revealed long-lived precipitation events in the heart
of the warm season, a period when rainfall is often
characterized as highly unpredictable or chaotic. The
coherency of these rainfall episodes suggests higher
predictability (i.e., beyond one or two days) than
previously thought possible.
Carbone, under the
auspices of the WMO World Weather Research Programme,
developed an international collaboration of scientists
spanning
five continents to explore the global significance
of recent warm season precipitation climatological
findings (Carbone et al. 2002). In contrast to the
U.S., where Doppler radar data were utilized on a continental
scale, geostationary satellite data are being applied
as a proxy. Participants in the multi-national collaboration
include Tai-Jen Chen (National Taiwan University, Republic
of China), Chung-Chieh Wang (Jin-Wen Institute of Technology,
Republic of China), Thomas Keenan (Australian Bureau
of Meteorology Research Center), Phillip Arkin
(University of Maryland), Vincenzo Levizzani and Laura
Zamboni (both of the University of Bologna, Italy),
and Arlene Laing (University of South Florida). Arkin
and Carbone are developing
a transfer function from the combined use of radar
and satellite data over the
continental U.S. This transfer function and adaptations
thereof will be applied to the satellite data statistics
obtained over East Asia (Chen and Wang), South America
(Arkin), Europe (Levizzani and Zamboni) Africa (Laing),
and Australia (Keenan). Early results from the geostationary
satellite data over East Asia and Australia confirm
a pattern of thermally-forced, continental-scale convection
that propagates across selected latitude bands during
the warm season of each region.
Data analysis software
In the customary case study approach, two-dimensional
images of radar data are examined frame by frame as
a function of time in order to understand the evolution
and dynamics of the event. To study the precipitation
cycle and understand its relationship to forcing mechanisms
over a long time series, techniques that reduce the
volume of data to a manageable level can provide more
insight. John Tuttle has
developed software to reduce the dimensions of the
dataset by averaging in one spatial dimension (latitude
or longitude) and plotting the data in the remaining
spatial dimension as a function of time (time-longitude
or time-latitude).
|
|
| Figure
11: Example
of a radar rainfall Hovmoller diagram for a strong
synoptically forced event over the U.S. The envelope
of precipitation associated with a low pressure
system can be seen moving eastward ~ 6 m/s. Embedded
within the envelope are streaks of precipitation
of shorter duration/length, propagating at 14 m/s.
These represent individual MCSs that initiated
within the envelope of favorable synoptic scale
forcing. |
In such plots (Fig. 11), coherent precipitation episodes
traveling across the continent appear as streaks, and
patterns associated with different types of forcing.
From the streaks, statistics characterizing the spatial-temporal
coherence and other properties of precipitation can
be calculated. For example, propagation speed, distance
(km) and duration (h) of individual events and their
probability density functions (PDFs) from an eight-year
period of record have been quantified. This software
has been employed by investigations worldwide in North
and South America, Asia, and Australia, and its use
will soon be expanded to Europe and Africa.
Cloud
microphysics and precipitation (top)
Convection initiation by colliding boundaries in the
United Arab Emirates
The United Arab Emirates has two distinct regions:
a large sandy desert along the southern edge of the
Arabian (Persian) Gulf comprising 90% of the country
and a spine consisting of the Hajar Mountains running
north to south along the eastern edge of the country
and the Gulf of Oman. These mountains peak at heights
of 1500 m and are broken by wadis (streambeds that
are usually dry except during the rainy season). Janice
Coen applied the Clark-Hall atmospheric model
to elucidate the complicated interactions
of a number
of local flow effects that occasionally combine to
produce preferred regions of
unusually heavy precipitation during the summer. A
simulation of the 31 July 2002 case is shown as an
example of this phenomenon. In the summer, the UAE
atmosphere is generally very dry and moisture comes
almost exclusively from the tropical monsoon flow which
the UAE is on the fringe of. This flow is heavily modified
by the Hajar Mountains and regularly produces orographically-induced
convective clouds in the afternoon as it passes west
over the Hajar Mountain range. The clouds, in turn,
produce evaporation-driven
downdrafts and streams of moisture that seep through
gaps, low-lying ridges, and wadis in the Hajar Mountains.
Meanwhile, the UAE also experiences an extremely strong
land-sea breeze where the heating of the land leads
to a strong, shallow northwesterly sea breeze that
may pushed ashore midmorning and extend 100 kilometers
inland. The collision of these flows (the moist easterly
gap flow and the northwesterly sea breeze, along with
the deep, well-mixed but extremely arid interior air)
first occurs in the northeast UAE, producing a bull's-eye
of converging flow, in this case, rainstorms near the
northeastern city of Dubai. As the gap flow strengthens
midday and the sea breeze penetrates further inland,
this bull's-eye location moves west over other coastal
cities. Thus, the net effect of these flows is to routinely
produce regular convective precipitation over the Hajar
Mountains. However, under unusual combinations of conditions
(related to the strength, direction, and moisture in
the monsoon flow), the flows may interact to produce
the rainstorms that can move west, away from the mountains
and toward the coastal cities.
|
|
| Movie 2: The Clark-Hall
atmospheric model was applied to elucidate
the complicated interactions
of a number of local flow effects that occasionally
combine to produce preferred regions of unusually
heavy precipitation during the summer, in the
United Arab Emirates. A simulation of the 31 July
2002 case is shown. |
Mouse over
image to begin movie. Alternately, you may download
the animation. |
Tropical cyclogenesis
Christopher Davis and
Lance Bosart (State University of New York, Albany)
continued their examination of tropical
cyclones during the 2000 and 2001 Atlantic hurricane
seasons, concentrating on late season storms forming
poleward of 20oN. All ten such storms featured a baroclinic
precursor, generally a sub-synoptic-scale trough that
had extruded deep into the subtropics and initiated
cyclogenesis, albeit very weak in some cases, along
a remnant frontal boundary. In all developing cases,
the vertical shear was initially near, or larger than,
the empirical threshold for tropical cyclogenesis;
in most cases, the shear decreased markedly prior to
the formation of a tropical storm. Davis and
Bosart hypothesized that in weakly baroclinic cases,
the pre-existing
disturbance focused convection that subsequently produced
a mesoscale vortex capable of self-amplification, similar
to Hurricane Diana (1984). In stronger cases, frontal
cyclogenesis provided the initial disturbance capable
of self-amplification,
through air-sea interaction. Through a detailed modeling
study of Hurricane Michael (2000), it was concluded
that diabatic transport and redistribution of potential
vorticity in the early convection was responsible for
the rapid decrease of vertical wind shear over the
nascent storm center.
Davis and Kate Musgrave, a SOARS student, investigated
the formation of Hurricane Gabrielle (2001), a storm
that formed in the northeastern Gulf of Mexico initiated
by a mid-tropospheric vortex from higher latitudes.
They hypothesized that the mid-tropospheric vortex,
embedded in weak but finite vertical wind shear, organized
convection that subsequently formed into Gabrielle.
The MM5 was used to investigate the basic formation
process and its dependence on the initial disturbance
and vertical wind shear. The control simulation realistically
simulated the formation of Gabrielle using a 12-km
grid spacing. Removing the initial disturbance using
potential vorticity inversion techniques resulted in
no storm being formed. Reducing the vertical wind shear
across the initial disturbance from about eight m/s
averaged through the troposphere to about four m/s
resulted in a marked slowing of cyclogenesis. This
is the first direct demonstration of a positive influence
of vertical wind shear in tropical cyclone formation
The interaction of inertia-gravity waves and balanced
flows
Improved prediction of precipitation, particularly
in the short range and at the mesoscale, depends in
part on predicting and understanding the occurrence
of inertia-gravity waves in mesoscale flows. For example,
the jet stream in the mid-latitudes is known from observations
to be an important source of inertia-gravity waves
(IGW). However, the mechanisms responsible for the
generation of IGW from balanced motions such as a jet
are as yet poorly understood, let alone quantified.
In collaboration with Hector Teitelbaum and Vladimir
Zeitlin (both from the Laboratoire de Meteorologie
Dynamique, France), Riwal
Plougonven (ASP) conducted
an analysis of inertia-gravity waves generated by the
jet over the North Atlantic, using observations (radiosondes)
collected during the Fronts and Atlantic Storm-Tracks
Experiment (FASTEX). Obtaining configurations of the
jet most favorable to gravity wave generation and detailed
case studies of wave generation events, the researchers
addressed the issue of how operational models describe
large-scale inertia-gravity waves that are under-resolved.
Possible applications for observational studies were
identified.
In collaboration with David Muraki (Simon Fraser University,
Canada), Chris Snyder and Plougonven investigated
the coupling of balanced motions and gravity waves
in a
shear flow. Idealized situations allow the analytical
quantification of the amplitude of the gravity waves
as a function of flow parameters. This relation was
investigated in more complex flow configurations with
numerical simulations using the WRF model. This fundamental
study contributes to the understanding of the relation
between balance and gravity waves in a shear flow.
Muraki and Snyder also
developed exact solutions for a vortex dipole in the
surface quasi-geostrophic equations
(Muraki and Snyder 2003). This dipole is a steadily
propagating, balanced, coherent structure that decays
rapidly with height. It represents an ideal starting
point for examining the interaction of balanced flows
with inertia- gravity waves since any emitted waves
will have clear signatures aloft owing to the decay
of the dipole.
Software development
William Hall and James
Dye developed analysis software
to examine in situ aircraft microphysical particle
data from both non-NCAR and NCAR data sets. The software
is available to the community at large and is being
applied to several project data sets including the
ABFM, IMPROVE II, CRYSTAL FACE, SHEBA, and UAE projects. Hall and
Masataka Murakami (Meteorological Research Institute,
Japan) put their microphysical scheme into
the Clark-Hall small scale dynamical model to include
additional microphysical variables and large scale
initialization options, providing an additional perspective
to modeling precipitation formation since the research
includes a wide range of meteorological conditions.
Precipitation formation in convective storms
Charles Knight continued to study first radar echoes
using the Severe Thunderstorm Electrification and Precipitation
Study (STEPS) data and emphasizing new information
that can be supplied by the differential reflectivity
(ZDR) in the early stages of precipitation. It is not
uncommon for there to be temporary, positive ZDR columns
to six or seven km MSL, indicating mm-sized water drops
being elevated to these levels before freezing, just
as the first precipitation echo is forming. The study
is incomplete, but so far this is seen particularly
in sheared convection, with the ZDR column at the upshear
side of the updraft, and generally separated from what
has been conventionally called the first precipitation
echo. The freezing of these drops may be important
in the spread of the ice phase in these early cumulus
clouds.
Ice-crystal Growth on Glass
Knight worked with Erick Adame, a visiting undergraduate
student in the SOARS program, on an experimental study
of ice growth on glass from the vapor at controlled
temperatures and supersaturations. The flowery, curving
growth of ice on glass has often been observed, but
never studied in any detail. The two main problems
are explaining the crystal orientations of the branching
growths and the (presumably systematic) curvature of
the crystal structure that accompanies the physical
curvature. The phenomenon turned out to be unexpectedly
complicated (much more so than snow), but systematic
features are emerging. It occurs to some extent in
ice growth on surfaces in general, but it is a new
crystal growth phenomenon, not yet understood, and
so it's potential importance is not clear.
|
