A VECTOR APPROACH TO FORECASTING MCS MOTION

1
An Introduction to Severe Weather Forecasting
NATIONAL SEVERE WEATHER WORKSHOP MIDWEST CITY,
OK 2 - 4 MARCH 2006 Updated Nov 2009 Stephen
Corfidi NOAA / NWS / NCEP Storm Prediction Center

2
Outline of Presentation

• Review basic thunderstorm structure and
  dynamics,
• identifying ingredients necessary for storm
  development
• Illustrate the severe weather forecast
  process, emphasizing ingredient
• evaluation using both observed and numerical
  model data
• Introduce selected tools, terms and processes
  commonly mentioned in
• severe weather discussions and forecasts
• Highlight some of the difficulties inherent to
  severe weather forecasting

Mulvane, KS, tornado, 12 June 2004 (J. Guyer)
3
Why forecast severe (convective) weather?

• The entire nation is affected throughout the year
  
• 400 deaths and 25 billion damage in U.S.
  annually
• 80 of presidential disaster declarations
  annually

Tri-State, 1925
Monticello, IN 1974

Waco, TX 1953

St Louis, 1896
(Statistics from Moller, 2001 Severe Local
Storms Forecasting, in Severe Convective
Storms, C. A. Doswell III, Ed., Amer. Meteor.
Soc.)
4
Severe convective weather includesmore than
tornadoes

• Hail (3/4 in. or greater)
• High Winds (58 mph or greater)
• Lightning (2nd leading killer)
• Flash flooding (1 killer
• 100 U.S. deaths annually)

(NOAA-NWS)
5
Thunderstorms
are the source of severe convective weather
The forecast task therefore reduces to

• Assessing the potential
• for thunderstorms

• Determining
• whether
• or not
• they will
• become
• severe

Above Towering cumulus near Everest, KS 10 June
1986 Left Cumulonimbus near Geneseo, KS, 1 June
2001
6
Storms require some combination of these three
ingredients

• Instability
• Lift
• Moisture

Instability Tendency for air parcels to move
up or down when displaced from rest
determined by rate of temperature change with
height Lift Mechanism(s) to
initiate, maintain or augment vertical air
motions (updrafts) Moisture Fuel in
the form of the latent heat of condensation (2.5
x 106 J/kg)
(McNulty, 1978)
7
Some storm Ingredient sources include
Instability (Thermodynamic) - Diabatic heating
(sun heats the ground) - Differential advection
(baroclinic environment) - Low-level moisture
inflow - Radiational cooling of cloud
tops Lift (Upward motion) - High level
divergence (jet streaks and associated
circulations) - Low level convergence - Low
level warm advection (isentropic ascent) -
Boundaries (fronts, dry lines, storm outflow, sea
breeze) - Terrain induced flow (upslope flow and
/ or obstacle effects) - Localized surface
heating - Gravity waves Moisture (Fuel in the
form of latent heat) - Evaporation /
Evapotranspiration - Pools left by previous
swaths of rainfall These processes occur on
many scales and are not mutually exclusive many
feedbacks and relationships are involved in
addition, the atmospheres response is often
non-linear
8
Life cycle of a thunderstorm cell the building
block of storms

(Doswell 1981 after Byers and Braham, 1949)
9
Animated life cycle of an idealized thunderstorm
cell
showing self-limiting nature of cells when winds
are weak and / or change little with height
rain-cooled storm outflow (blue) undercuts the
updraft that gave birth to the cell

(Loop courtesy of NWS-Tulsa)
10
Thunderstorm cells are rarely isolated more
often they occur in the form of MULTICELL
CLUSTERS



Radar depiction
Nearby side view

Time sequence, front view
Distant front view
(Sequence adapted from NOAA / Doswell 1985)
11
Sustained thunderstorms require one additional
ingredientVertical Shear

...the change in wind speed and
/ or direction with height

Developing storm with shear-tilted updraft,
Baltimore, MD, 13 July 1977
12
Sheared environments foster the development of
sustained storms
When winds increase strongly and / or change
direction with height, updraft (red arrow) is
slanted rain-cooled storm outflow (blue) can
then spread downshear away from updraft storm
therefore avoids tendency to self-destruct
10 km
8 km
6 km
4 km
2 km
2-10 km Shear
(Moller et al. 1994 after Browning and Ludlam,
1962)
13
Helping separate updraft and downdraft is
important, but the most significant influence of
vertical shear on a storm is in fostering
mid-level storm rotation The presence of a
Mesocyclone (or Meso) dynamically
enhances updraft steadiness and longevity
Storms with deep, persistent mesocyclones are
called Supercells
Animation of Klemp 1987 (UCAR / COMET)
Colorado supercell (G. Moore)
14
Early aerial view of a supercell by T.
Fujita Looking northeast, showing southeasterly
low level winds (green) veering to southwesterly
(yellow), and then strong west northwesterly
(blue) with height
Near Topeka, KS 21 April 1961 (SMRP Research
Paper 16)
15
Supercells dont always occur singly but often
in lines and clusters when the lifting agent is
linear, such as along a front or thunderstorm
outflow boundary
Squall line shelf cloud over Ventnor, NJ 24 July
1979
The severe weather hazards most likely to occur
--- and their longevity --- are not only a
function of storm type, but also storm
arrangement, i.e., the Convective Mode
Radar and satellite views of TX / OK squall
line, 4 March 2004
16
But not all severe weather is spawned by
supercells
downbursts, damaging straight-line winds,
landspouts, and flash flooding can and / or
often do occur without supercells
17
The Forecast Process
Forecast Diagnosis Trend Diagnosis
requires good analysis (more on this
later) Trend is determined via 1.
Extrapolation 2. Climatology 3.
Forecaster Knowledge a. Pattern recognition
( gt Conceptual models) b. Ingredient
evaluation ( gt Composite charts) 4.
Numerical model guidance (both synoptic and
mesoscale, and statistical applications used to
enhance model output, such as ensemble
techniques)
Maryland pulse storm, 13 July 1977
Oklahoma pulse storm, 13 Sep 1997
(Doswell 1986 Moller 2001)
18
The relative value of different methods used to
determine trend varies over forecast range...
Only curve controlled by forecasters
Relative utility
Time (hours)
(Adapted from Doswell, 1986 Short-range
Forecasting)
19
Climatology provides a start

• Most useful when responsible physical
  processes are understood
• Can provide a useful first guess when small
  scale forcing mechanisms are not
• known or are unresolvable

4 March

8 April
Probability of an F2 or stronger tornado within
25 miles of a point
6 May
(From NSSL http//www.nssl.noaa.gov/hazard/)
20
Numerical Model Guidance

• Model output is indispensible to forecasting
  weather forecast models
• are an under-appreciated achievement of modern
  science
• Models translate atmospheric physics into
  numbers
• Successful use requires basic knowledge of a
  models
• - Finite differencing methods (numerical
  approximations
• to the real world)
• - Data assimilation techniques (what data go in
  how are
• they treated - - - more is not necessarily
  better!)
• - Parameterization of physical processes
  (especially boundary
• layer, convective and radiation effects)
• Numerical models remain imperfect
• - Effects of convection / convective outflow
  (cold pools)
• - Boundary layer processes (e.g.,
  evapotranspiration
• turbulent transfer of moisture / heat /
  momentum)
• - Data irregularities in space and time (e.g.,
  surface obs
• every few minutes raobs only twice daily)

21
The ability of a model to predict a phenomenon is
scale-dependent

Convergence area (shaded) is just minimally
resolved by the grid of this hypothetical model
(4 grid box widths required) The individual
thunderstorms shown are too small to resolve
Tornado watch
(Adapted from David Bright, NOAA-SPC)
22
Resolution affects the value of model guidance
Output that may be very useful for convective
outlook generation (i.e., for the synoptic or
large scale) may be of limited use for watch
preparation (i.e., for smaller space and time
scales)

A typical SPC severe weather outlook and watch
23
A models poor precipitation forecast can
negatively affect its depiction of instability
and convective evolution


Forecast total precipitation
Forecast convective precip. w/ lapse rate

Observed (radar) squall lines
The Super Outbreak, 00Z 4 Apr 1974 148
tornadoes in 24 hrs
Model used A 29 km, 50 layer version of the Eta
(similar to that used operationally the late
1990s)
24
while at the same time (sometimes!) still
providing a good 36-hr forecast of surface
pressure (left) and mid-level flow (right)


Observed
Observed
Forecast
Forecast
25
Numerical Guidance Data display
Viewing several atmospheric levels simultaneously
aids in applying conceptual models

Pcpn / Vert Motion
Sfc Pres / Thkns
500 mb Hgt / Temp
850 mb Hgt / Temp
26
Numerical Guidance Ensemble forecasts
Ensemble forecasts are a set of forecasts all
valid at the same time, but made using (1)
different models and / or (2) the same model with
different initial conditions, parameter settings,
etc.

Ensemble probability of 3-hr precip gt .01 15 hr
fcst, valid period 2100Z 31 Aug to 0000Z 1 Sept
2004
Ensemble mean precip 0.01 (thick dashed)
27
Numerical Guidance Ensemble forecasts (contd)
Cloud Physics Thunder Parameter (CPTP) was
developed to refine thunderstorm forecasts based
on more traditional model-generated parameter
fields such as CAPE

Ensemble probability of CPTP gt 1 15 hr fcst,
valid period 2100Z 31 Aug to 0000Z 1 Sept 2004
- Sufficient CAPE in the 0o to -20o C layer -
Lifting condensation level gt -10o C -
Equilibrium level temperature lt -20o C
Ensemble probability of CPTP gt 1 (thick dashed)
28
Numerical Guidance Composite ensemble forecasts
A composite ensemble forecast of thunderstorm
potential it is the product of the two previous
ensemble forecasts

Ens Prob (Precip gt .01) x Ens Prob (CPTP gt 1) 15
hr fcst, valid period 2100Z 31 Aug to 0000Z 1
Sept 2004
29
Numerical Guidance Calibrated ensemble
forecasts
Composite model forecast of the previous slide
has been calibrated (i.e., modified) by
performance of that same parameter in recent
weeks (see Bright et al. 2004)

Calibrated ensemble tstm parameter 15 hr fcst,
valid period 2100Z 31 Aug to 0000Z 1 Sept 2004
Observed lightning strikes (yellow crosses)

30
Conceptual Models
provide a good starting point for the forecaster
L
L
L
Idealized 36-hr evolution of a mid-latitude
synoptic scale surface low (After Palmen and
Newton, 1969)
Idealized 3-D warm front, showing comparatively
shallow upglide (isentropic ascent) above
frontal surface
Strength, speed and track of surface low
dependent upon jet stream pattern aloft
L
Idealized 3-D cold front, showing thunderstorms
along leading edge of front
L
(Color figures Golden Guide to Weather)
31
Conceptual models aid diagnosisbut must be used
carefully
L
NY
PA
OH
KY
High wind reports
Idealized setup of synoptic-scale features
associated with a classic tornado outbreak (After
Barnes and Newton, 1983)
Pressure / frontal pattern of analysis (right)
resembles set-up of a classic tornado outbreak
(above) But moisture / temperature fields are
not supportive of discrete supercells
Hand-drawn surface analysis associated with a
cool-season forced ascent convective band (Ohio
Valley, 9 March 2002)
32
Conceptual models must be updated and linked
as experience and physical understanding grow

ELONGATION OF A COLD POOL IN UNIDIRECTIONAL FLOW
Concurrent bow echo / flash flood MCS
Derecho / Bow Echo MCS Pattern (Johns 1984)

FLASH FLOOD MCS
DERECHO MCS
Synoptic type flash flood pattern (Maddox 1979)
33
Composite Chartscombine information derived
from observations and numerical guidance with
conceptual models
Emphasis is on juxtaposition of storm
ingredients, storm limiting factors (e.g.,
caps), and their 4-D evolution
L
Storm reports 24-hr period 4-5 May 2003
24-hr composite prog valid 0000Z 5 May 2003
34
Diagnosis The First Step in Forecasting

• Need to know how the atmosphere reached
  current state before a
• forecast can be made
• Successful diagnosis requires good analysis
  poor analysis
• reduces forecast accuracy
• Good analysis, in turn, requires
• - Understanding of on-going 4-D (space-time)
  interactions between the synoptic -, meso -,
  and storm-scales
• - Knowledge of topography
• - Experience (pattern recognition / conceptual
  models)
• - Ability to synthesize data from multiple
  sources (e.g., surface observations and spotter
  reports with radar, satellite and model data)
• - Continuity (a single snapshot is not
  sufficient)

Goal is to accurately assess storm ingredients
35
Sunday 4 May 2003 High Risk over the Central
Plains
7 am CDT An intense upper level trough and
120 kt speed max over the southwest U.S.
(left) is poised to overspread a band of very
moist, southerly low level flow from the
Gulf of Mexico (below)
L
500 mb Mid Level (about 3 miles altitude)
250 mb Upper or High Level (about 6 miles
altitude)
WIND STRENGTH NOTATION Two flags 100 kts Flag
50 kts Bar 10 kts Half bar 5 kts
L
850 mb Low Level (about ¾ miles altitude)
36
Sunday 4 May 2003 High Risk over the Central
Plains
7 pm CDT The high level speed max has
moved east and nearly overtaken an
intensified low level circulation now over
southwest Iowa a major tornado outbreak
has been in progress since 4 pm in eastern
Kansas and western Missouri
L
500 mb Mid Level (about 3 miles altitude)
250 mb Upper or High Level (about 6 miles
altitude)
WIND STRENGTH NOTATION Two flags 100 kts Flag
50 kts Bar 10 kts Half bar 5 kts
L
850 mb Low Level (about ¾ miles altitude)
37
Sunday 4 May 2003 High Risk over the Central
Plains
Surface evolution A strong low pressure center
(L) moves from northwest Kansas to
northwest Missouri, accompanied by a
well-defined warm front, dry line and cold
front
1200Z (700 am CDT)
1800Z (100 pm CDT)
0000Z (700 pm CDT)
(Adapted from NCEP / HPC)
38
Sunday 4 May 2003 High Risk over the Central
Plains
Satellite loop, sunrise to sunset Strong
storms first form around midday near
surface low in southern Nebraska by mid
afternoon, other more significant tornadic storms
develop south along dry line over eastern
Kansas later still, additional severe
storms form on cold front in north central Kansas
Storm reports 24-hr period, 4-5 May 2003

39
Sunday 4 May 2003 High Risk over the Central
Plains

1445Z
40
Sunday 4 May 2003 High Risk over the Central
Plains

1545Z
41
Sunday 4 May 2003 High Risk over the Central
Plains

1645Z
42
Sunday 4 May 2003 High Risk over the Central
Plains

1745Z
43
Sunday 4 May 2003 High Risk over the Central
Plains

1845Z
44
Sunday 4 May 2003 High Risk over the Central
Plains

1945Z
45
Sunday 4 May 2003 High Risk over the Central
Plains

2045Z
46
Sunday 4 May 2003 High Risk over the Central
Plains

2145Z
47
Sunday 4 May 2003 High Risk over the Central
Plains

2245Z
48
Sunday 4 May 2003 High Risk over the Central
Plains

2345Z
49
Sunday 4 May 2003 High Risk over the Central
Plains

0045Z
50
Analysis of sounding (RAOB) data


Balloon and equipment package ascending (at rate
of about 1000 per minute)


Equipment package

Radiosonde release (Package can reach a height of
115,000 and can travel more than 125 miles from
release location)
Radiosonde shelter
Twice-daily raob release sites ( 70 in U.S.)
(NOAA NWS)
51
Unabridged RAOB data display

Temperature
Dewpoint
Wind profile Flag 50 kts Bar 10 kts Half bar
5 kts
200 mb (High level)
Parcel ascent
Hodograph (another way to display wind profile)
Temperature (0 o C)
500 mb (Mid level)
Moisture
850 mb
For convenience, radiosonde observations
typically are evaluated separately for
thermodynamic (stability) and kinematic (wind)
aspects
Surface
Wind (kinematic) parameters
Thermodynamic parameters
52
RAOB Analysis Thermodynamic datausing a Skew
T - Log P chart

Environmental Temperature
Environmental Dewpoint
Temperature (-20 o C)
200 mb (High level)
Parcel ascent (if saturated)
Temperature (0 o C)
500 mb (Mid level)
Dry Adiabat
Temperature (20 o C)
An advantage of Skew T Log p charts is that the
temp. lines cross the dry adiabats at a large
angle, making lapse rates easy to diagnose
Drier Moister
Moisture
850 mb
Surface
53
RAOB Analysis Thermodynamic data Skew T - Log
P usage is based on Parcel theory


Environmental Temperature
Equilibrium level
Environmental Dewpoint
Cumulus clouds with bases at LCL / LFC
Parcel ascent (if saturated)
Parcel theory based on numerous assumptions
ignores entrainment of environmental air
Lifting Condensation Level and Level of Free
Convection (not always equivalent)
(Diagram NOAA NWS)
54

Parcel ascent curve vs. Environmental temperature

13o C temperature difference
Temperature difference between parcel and
environment at 500 mb is known as the Lifted
Index

5o C temperature difference
Parcel ascent is everywhere to the left of
temperature curve atmosphere is stable to
vertical displacement
Updraft strength is directly proportional
to the difference between the parcel ascent
curve (dashed white line) and the
environmental temperature profile (solid
red)
55
CAPE Convective Available Potential Energy

CAPE (yellow)

CAPE is the total area between the parcel ascent
curve and the environmental temperature profile
CAPE is the total energy available for
thunderstorm updrafts
CAPE (yellow) with CIN (blue)

Layer of very steep lapse rates (EML)
No CAPE all CIN
Convective inhibition (CIN) is area to the
right of parcel ascent curve when CIN is
present simultaneously with CAPE, it acts
to cap convective development (middle
example above)
56
RAOB data Comparing Skew T plots with the real
world
LCL - Lifting condensation level LFC - Level
of free convection EL - Equilibrium level




Flat cumulus streets beneath a cap
Buoyant, cumuliform structure above LFC

Laminar cloud structure between LCL and cap
Anvil top forms at equilibrium level (EL)
Near Anadarko, OK 3 April 2003 (V. Doswell)
Laminar cloud structure between LCL and cap
SW Oklahoma, 22 May 1995
SE Nebraska, 20 April 1985
57
Downdrafts operate differently from updrafts

• Storm updrafts owe their buoyancy to
  condensation, but strong downdrafts
• may be saturated or unsaturated
• Downdrafts begin with negative buoyancy, but
  drag effect due to precipitation
• also very important
• Small precipitation particles hasten
  evaporation, sublimation and melting ---
• and, therefore --- cooling rate in descending
  parcel
• In short, strong downdrafts require environmental
  lapse rates sufficiently steep to overcome
  compressional warming and / or assistance from
  processes occurring inside the parcel
  (evaporation, etc.) to offset warming due to
  sinking

Inverted Vee sounding characteristic of
downburst events


(Downburst evolution Bill Bunting)
58
RAOB sequence 1 Thermodynamic evolution at
Topeka, KS

0000Z Sunday 4 May 2003

Minimal elevated CAPE
59
RAOB sequence 1 Thermodynamic evolution at
Topeka, KS

1200Z Sunday 4 May 2003

CAPE has increased as lower levels
have moistened, but instability is still
elevated
60
RAOB sequence 1 Thermodynamic evolution at
Topeka, KS

1800Z Sunday 4 May 2003

Daytime heating and continued low level moisture
inflow have boosted CAPE to more than 4500 J kg-1
Kansas City KS MO F4 tornado occurred
approximately 3 hrs later, 50
miles east of raob site
Instability is now surface-based
61
RAOB sequence 1 Thermodynamic evolution at
Topeka, KS

0000Z Monday 5 May 2003

CAPE nearly disappears as dry air moves inbut
low to mid level lapse rates have steepened
Deeply-mixed boundary layer with very steep lapse
rates (rapid temperature decrease with height)
62
RAOB sequence 2 Thermodynamic evolution at
Springfield, MO

0000Z Sunday 4 May 2003

Minimal, very elevated CAPE
63
RAOB sequence 2 Thermodynamic evolution at
Springfield, MO

1200Z Sunday 4 May 2003

CAPE has increased as lower levels
have moistenedbut as at Topeka at 1200Z,
instability remains elevated
64
RAOB sequence 2 Thermodynamic evolution at
Springfield, MO

1800Z Sunday 4 May 2003
CAPE increase has been stymied by arrival of
EML (elevated mixed layer with steep lapse
rates) The EML effectively caps the lower
atmosphere to deep convectiondespite fact that
air near the ground has continued to warm and
moisten CAPE therefore remains primarily
elevated

Classic loaded gun sounding (Fawbush and Miller
1953)
65
RAOB sequence 2 Thermodynamic evolution at
Springfield, MO

2000Z Sunday 4 May 2003
Surface heating, low level moisture inflow
and mesoscale ascent have eroded most of
cap Sounding is now nearly supportive of
surface-based storms, with only
minimal additional lifting needed

66
RAOB sequence 2 Thermodynamic evolution at
Springfield, MO

0000Z Monday 5 May 2003

Cap completely eliminated
Substantial surface-based CAPE now present, with
low LCL
Pierce City Battlefield, MO F3 tornado
in progress at this time, lt 40 miles
southwest of raob site
67
(Back to our) Unabridged RAOB display

Temperature
Dewpoint
CONVENTIONAL WIND PROFILE Flag 50 kts Bar 10
kts Half bar 5 kts
200 mb (High level)
Parcel ascent
HODOGRAPH (another way to display wind profile)
500 mb (Mid level)
850 mb
Surface
Wind (kinematic) parameters
Thermodynamic parameters
68
RAOB analysis Focus on kinematic datausing
either conventional wind profiles or hodographs

HODOGRAPH Provides better visualization of
changes in both wind direction and speed with
height
CONVENTIONAL WIND PROFILE Flag 50 kts Bar 10
kts Half bar 5 kts
15 km
1 km
200 mb (High level)
12 km
2 km
3 km
9 km
8 km
6 km
Surface
9 km
500 mb (Mid level)
6 km
3 km
700 mb
850 mb
1 km
Surface
69
The Hodographa graphical tool used to evaluate
vertical wind shear

Vector length is proportional to wind speed
direction parallel to direction of flow
Wind vectors drawn from the origin in the
direction toward which the wind is blowing
To create a hodograph, the wind vectors (red) are
first plotted on a polar coordinate chart On a
polar chart, the axes represent the four cardinal
compass directions the concentric circles
depict wind speed (here, in knots) Usually, the
entire wind vectors are not drawn, only their end
points the hodograph (yellow curve) is
then plotted by connecting the ends of each wind
vector Recall that vertical wind shear is the
change in wind with height thus, each
segment of the hodograph depicts the shear
(i.e., speed and direction) in that layer
(Adapted from UCAR / COMET)
70
The Hodographadditional examplesshowing how
different wind profilesyield wide variationsin
hodograph shape


STRAIGHT HODOGRAPH wind direction remains
constant, but speed increases with height

STRAIGHT HODOGRAPH even though both wind
speed and direction change with height
Hodograph is important because storm type,
motion and evolution are all influenced by the
vertical shear
CURVED HODOGRAPH wind speed remains
constant, but direction changes with height
(Adapted from UCAR / COMET)
71
Recall origin of supercell development with a
straight hodograph
Example application of hodograph data
Right-mover
Left- mover
Time To
Time To T
1. Horizontal vorticity associated with shear
of the ambient wind is tilted into the
vertical by storm updraft, yielding mirror-
image rotating updrafts (mesocyclones
red and blue arrows) 2. The mesocyclones are
stongest at mid levels as this is where
updraft and, therefore, stretching, are
typically greatest (recall soundings
showing level of maximum CAPE)
3. Mid level rotation with the mesocyclones
causes pressures to fall aloft (roughly 2-3
miles above ground) 4. The pressure falls
further encourage original storm to split
into paired right- and left- moving
members 5. Neither member, however, is favored
for growth by the updraft blocking
effect, as influence of the latter occurs
on the upshear and downshear sides of the
the original storm (shaded arrows above)
(Adapted from Klemp 1987)
72
When the hodograph curves clockwise with height,
the right-moving storm is favored for sustained
growth / longevity
Application of hodograph data (contd.)
STRAIGHT HODOGRAPH
CURVED HODOGRAPH
Right-mover
Left- mover
Same diagram as in previous slide, showing
mesocyclones on right and left flanks of
original storm, 90o away from forcing
associated with updraft blocking effect
(shaded arrows)
When winds and hodograph turn clockwise with
height, the updraft blocking effect
results in an upward-directed pressure force
that favors growth of the right-moving cell
(red arrow), and works against the left-
mover (blue arrow)
(Adapted from Klemp 1987)
73
RAOB Analysis Kinematic dataComparison of
conventional wind profile to hodograph

HODOGRAPH
CONVENTIONAL WIND PROFILE Flag 50 kts Bar 10
kts Half bar 5 kts
15 km
1 km
200 mb (High level)
12 km
2 km
3 km
9 km
8 km
6 km
Surface
9 km
500 mb (Mid level)
6 km
3 km
700 mb
Topeka, KS 1200 Z Sun. 4 May 2003
850 mb
1 km
Surface
74
RAOB Analysis Kinematic dataConnecting tips
of plotted wind vectors to yield hodograph curve

HODOGRAPH
CONVENTIONAL WIND PROFILE Flag 50 kts Bar 10
kts Half bar 5 kts
15 km
1 km
200 mb (High level)
12 km
2 km
3 km
9 km
8 km
6 km
Surface
9 km
500 mb (Mid level)
6 km
3 km
700 mb
850 mb
1 km
Surface
75
RAOB Analysis Kinematic dataPlotting observed
(or estimated) storm motion

HODOGRAPH
CONVENTIONAL WIND PROFILE Flag 50 kts Bar 10
kts Half bar 5 kts
15 km
1 km
200 mb (High level)
12 km
2 km
3 km
9 km
8 km
6 km
Surface
9 km
500 mb (Mid level)
6 km
Storm motion 237o / 35 kts
3 km
700 mb
850 mb
1 km
Surface
76
RAOB Analysis Kinematic DataDetermining
storm-relative flow

CONVENTIONAL WIND PROFILE Flag 50 kts Bar 10
kts Half bar 5 kts
HODOGRAPH
15 km
1 km
200 mb (High level)
12 km
2 km
3 km
9 km
8 km
6 km
Surface
9 km
500 mb (Mid level)
6 km
Storm motion 237o / 35 kts
3 km
700 mb
Blue lines depict storm- relative
motion at the surface and 1 km levels

850 mb
1 km
Surface
77
RAOB Analysis Kinematic DataGraphical
depiction of Storm-relative helicity (SRH)

CONVENTIONAL WIND PROFILE Flag 50 kts Bar 10
kts Half bar 5 kts
HODOGRAPH
15 km
1 km
200 mb (High level)
12 km
2 km
3 km
9 km
8 km
6 km
Surface
9 km
500 mb (Mid level)
6 km
Storm motion 237o / 35 kts
3 km
700 mb
Area swept out by the blue lines is
proportional to the Storm-Relative
Helicity (SRH) in the surface-to-1 km
layer
850 mb
1 km
Surface
78
RAOB Analysis Kinematic DataGraphical
depiction of Storm-relative helicity (SRH) Part
2

CONVENTIONAL WIND PROFILE Flag 50 kts Bar 10
kts Half bar 5 kts
HODOGRAPH
15 km
1 km
200 mb (High level)
12 km
2 km
3 km
9 km
8 km
6 km
Surface
9 km
500 mb (Mid level)
6 km
Storm motion 237o / 35 kts
3 km
700 mb
Similarly, area swept out by the outer
two blue lines is proportional to the
SRH in the surface-to- 3 km layer
850 mb
1 km
Surface
79
RAOB Analysis Kinematic DataSummary of
hodograph analysis

CONVENTIONAL WIND PROFILE Flag 50 kts Bar 10
kts Half bar 5 kts
HODOGRAPH
Hodograph analysis allows for rapid estimation of
SRH and, therefore, the likelihood for
tornadoes But utility of SRH data is dependent
on accuracy of estimated or observed storm
motion Storm motion is complex, based on the
environmental wind field (speed and shear), and
other external factors (such as the
distribution of instability and the location of
fronts / outflow boundaries, etc.) These
factors change over space and time!
15 km
1 km
200 mb (High level)
12 km
2 km
3 km
9 km
8 km
6 km
Surface
9 km
500 mb (Mid level)
6 km
Thus, prudent hodograph analysis must be
accompanied by concurrent evaluation of
surface, radar and satellite data
Storm motion 237o / 35 kts
3 km
700 mb
Similarly, area swept out by the outer
two blue lines is proportional to the
SRH in the surface-to- 3 km layer
850 mb
1 km
Surface
80
RAOB Sequence 3 Kinematic evaluation at
Springfield, MO

Substantial (gt 50 kts) deep shear already
present Hodograph is almost unidirectional
despite curved wind profile
0000 Z Sunday 4 May 2003
81
RAOB Sequence 3 Kinematic evaluation at
Springfield, MO

Strengthening low to mid level winds associated
with LLJ and approach of upper trough have
contributed to an increase in both deep shear and
SRH
Not all of the SRH is, however, effective, as
any storm at this time is likely to be elevated
1200 Z Sunday 4 May 2003
82
RAOB Sequence 3 Kinematic evaluation at
Springfield, MO

Deep shear slightly weaker now due to temporal
backing of winds that has occurred at mid
levels Low to mid level flow remains very
strong, and lower part of hodograph has filled
out
But effective SRH has been diminished by arrival
of elevated mixed layer and its associated cap
1800 Z Sunday 4 May 2003
83
RAOB Sequence 3 Kinematic evaluation at
Springfield, MO

Cap has been eliminated, and instability now
nearly surface-based Deep shear remains more
than sufficient for sustained supercells
0 1 km SRH has increased (in this case mainly
due to faster estimated storm motion) to nearly
500 m2 s-2, and is now largely effective
2000 Z Sunday 4 May 2003
84
RAOB Sequence 3 Kinematic evaluation at
Springfield, MO

Both kinematic and thermodynamic setups now very
favorable for supercells and tornadoes
0000 Z Monday 5 May 2003
85
RAOB data do not tell the whole storystorm mode
must be considered when analyzing severe threat
Birmingham, AL 18z 16 Feb 2001
Birmingham, AL 18z 16 Dec 2000
CAPE 1194 J kg-1 CIN 0 J kg-1 SRH 120 m2
s-2 Shear 68 kts
CAPE 1351 J kg-1 CIN 2 J kg-1 SRH 213 m2
s-2 Shear 73 kts
Tornado and high wind reports
Quasi-discrete supercells
Bow echo / derecho
High wind reports
86
Subjective Surface AnalysisWhy it is essential
to severe weather forecasting

• Surface observations are a rich data source
  much denser in
• space and time than raob data
• Facilitates mental process of synthesizing
  data from various platforms
• Helps one develop valid conceptual models
• Necessary for tracking motion / intensity of
  mesoscale features
• such as outflow boundaries, sea breeze fronts,
  etc.
• Provides a reality check serves to keep the
  forecaster ---
• and computer analyses / forecasts --- connected
  to the real world

Surface analysis must be performed regularly to
maintain skill and to successfully track subtle
features
87
Subjective analysis helps diagnose a
poorly-forecast boundary and surface wavea
system that was responsible for 50 tornadoes,
including 2 F3s
21Z 27 July 1994
00Z 28 July 1994
03Z 28 July 1994
88
Boundaries are amongst the most
importantfeatures to be diagnosed by subjective
analysis
Why boundaries are important

• Often a location for convective initiation
• Are regions of enhanced low-level shear
• Mark axes of enhanced horizontal vorticity
• that can be tilted vertically by updrafts
• to foster supercell development and,
• therefore, storm strength / longevity
• Tornadic storms often confined to
• boundaries when background SRH is low
• Convective mode is, to a great extent, a
  function of
• boundary orientation and type (e.g., Birmingham,
  AL
• case previously shown also, elongating gust
  front
• conceptual model shown earlier)

Flash flood storm on outflow boundary, Glenarm,
MD, 7 Aug. 1978
89
Boundary detection is facilitated by data
synthesis
All data from c. 2230Z Sunday 22 June 2003
Record 6 - 7 diameter hail In Aurora, NE
(yellow square)
BASE REFLECTIVITY
Outflow boundaries
Thermal axis
Moisture axis
MESOANALYSIS
Also useful lightning, profiler and ACARS data
VISIBLE SATELLITE
90
and by use of extended animations
Base reflectivity radar loop , Hastings, NE 2104Z
Sunday 22 June 0528Z Monday 23 June 2003
Visible data satellite loop, 1322Z Sunday 22
June 0320Z Monday 23 June 2003 Loop shows
MCSs (storm clusters) Supercell formation Outflow
boundaries Cloud streets Wave clouds Gravity
waves Dry line
91
Objectively-derived fields provide valuable
assistancewith subjective analysis

• Computer-drawn streamlines help locate fronts
  and other
• potential sites of storm development
• Animation of objectively-drawn derived fields
  helps forecaster
• spot influence of bad data, especially when
  displayed
• with conventional data
• Combination of computer and human skills
  maximizes
• analysis skill and, therefore, forecast
  potential

92
In turnconfidence in objectively-derived
fieldsmay be enhanced with conventional data
Shape / orientation of SIGTOR axis validated by
independent confirmation of outflow boundary
location in satellite imagery
SIGTOR Parameter
2100Z
Storm rpts
2200Z
Visible data satellite
2300Z
93
Acknowledgements Pete Banacos David
Bright Chris Broyles Greg Carbin Sarah
Corfidi Jared Guyer Jason Levit Dan McCarthy Jeff
Peters Jason Ribelin Russ Schneider Steve Weiss

Retreating NW flow storm near Lindsborg, KS, 1
June 2001